Durgical
EN
Open the app

Using the electronic record safely

  • Concept

The electronic record makes it easy to write a lot, order fast and never look again. Each shortcut is a known way patients get harmed.

Updated

What it is

The electronic health record removes old errors (illegible writing, lost charts) but brings its own: copied text that is no longer true, orders placed on the wrong patient, alerts nobody reads, templates that record findings never examined and results nobody follows up. Safe use means habits and system design that keep the record true, the order on the right patient and every result closed.

Key facts

  • Copy-paste safe practice: copied material should be identifiable, its source traceable, staff trained, and its use monitored.
  • Wrong-patient orders: several open records, similar names and interruptions are the usual setting; one record at a time and a visible photograph or identifiers help.
  • Templates and defaults: a pre-ticked «normal» examination documents findings that were never checked and hides the reasoning.
  • Results follow-up: «acknowledged» in an inbox is not the same as acted on; abnormal results need a closed loop with the patient.
  • A good note: problem-oriented, with subjective, objective, assessment and plan, where the assessment states what you think and why.
  • Downtime: paper forms, a read-only copy or printed summary of current patients, two-identifier checks and later back-entry of everything done on paper.
  • Voice recognition (speech-to-text): it mishears sound-alike words, drug names, numbers and small words such as ‘no’ or ‘not’, so a dictated note must be read and corrected before it is signed.

Worked example

  1. A ward audits 40 progress notes from one month: 28 contain text copied from earlier days (28 / 40 = 70 %).
  2. Of the 40 notes, 6 repeat a finding that was no longer true, such as a resolved fever: 6 / 40 = 15 % of notes carry an outdated fact.
  3. The hospital places 100 000 orders a year; at about 58 wrong-patient orders per 100 000, that is about 58 caught near misses.
  4. An identity re-entry step lowering the odds by 41 % would leave about 58 × 0.59 ≈ 34 a year, at the cost of a few seconds per order session.
  5. The ward adds a rule: write the assessment fresh each day and mark copied text; a repeat audit finds 1 outdated finding in 40 notes (2.5 %).
  6. Reading: the record became shorter and truer without banning the tool; the design and the habit changed together.

Electronic record hazards and their barriers

HazardHow it harmsBarrier
Copy-pasteOutdated or wrong facts look current; notes too long to readWrite the assessment fresh; mark and attribute copied text
Wrong-patient orderMedicine or test for the wrong personOne record open at a time; check two identifiers; re-entry or photo prompts
Alert fatigueThe important alert is dismissed with the trivial onesFewer, tiered, non-repeating alerts
Default templatesFindings recorded that were never examinedRecord only what was done; avoid pre-ticked normals
Unclosed resultsAn abnormal result never reaches the patientNamed owner, tracked follow-up, patient informed
DowntimeOrders and results lost while the system is downRehearsed paper procedure and back-entry

In practice

  • Before signing, read the note as the next clinician would: is every line true today, and is your reasoning visible?
  • Place orders with only one patient's record open, and check the name and a second identifier on the order screen.
  • Treat an alert you override as a decision: if it was right to override, note why in a few words.
  • Know your service's downtime kit and who back-enters paper records once the system returns.

Common pitfalls

  • Believing that more text is safer: long copied notes hide the one change that matters.
  • Fixing an old error by deleting it: corrections are added with date and author, never erased.
  • Assuming the system will catch a wrong-patient order: most are caught only by the person ordering.
  • Signing a result as seen without a plan for it.

Common traps

  • Copy-paste is forbidden: it is allowed and common; the risk is copying without checking, so copied text must be verified, marked and kept short
  • A longer note protects me legally: a bloated note with contradictions shows that it was not read; a short, true, reasoned note protects patient and doctor
  • If the system raised no alert, the order is safe: alerts miss many errors, and most wrong-patient orders are caught only by the prescriber
  • Acknowledging a result in the inbox closes it: the loop closes only when the result is acted on and the patient knows
  • A complete template means a complete examination: pre-ticked normals record findings that may never have been checked

Test yourself

Think of the answer, then tap the question to see it.

Day 5 of an admission: your note, copied from day 2, still says «febrile». The patient has been afebrile for 48 hours. What is the risk?

The next clinician may trust a false finding, for example prolonging antibiotics. Write today's findings fresh and correct the copied text.

You have three records open and are interrupted while ordering. Which error is most likely and how do you prevent it?

A wrong-patient order. Close the other records, return to one patient and check name and a second identifier before signing.

A potassium result of a discharged patient arrives abnormal. Who is responsible for acting on it?

The clinician or team who ordered it, until it is acted on and the patient is told; acknowledging it in the inbox is not enough.

Why does a pre-ticked «normal» examination template threaten safety?

It records findings that may not have been examined and hides the reasoning, so a real abnormality can look ruled out.

Add to my cards (4)

Elsewhere in DurgiPedia

Connections

← All of epidemiology

Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.

Sources: Safe Practices for Copy and Paste in the EHR. Systematic Review, Recommendations, and Novel Model for Health IT Collaboration (Applied Clinical Informatics (PubMed Central), 2017); Understanding and preventing wrong-patient electronic orders: a randomized controlled trial (Journal of the American Medical Informatics Association (PubMed Central), 2013); Effects of workload, work complexity, and repeated alerts on alert fatigue in a clinical decision support system (BMC Medical Informatics and Decision Making (PubMed Central), 2017).

Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake

Representative model, for educational purposes only.