Before You Sign: Reviewing an AI-Drafted Allergy Note

The short answer: an AI-drafted note is a draft until the clinician

signs it. Before signing an allergy note, check six things: the allergen

and reaction history, the test results, the immunotherapy details, the

pertinent negatives, the codes on each assessment item, and whether the

plan matches what the patient was told. Check them against the source,

not against how confident the draft sounds.

What a new study does and doesn't tell us

A quality-improvement study published in Annals of Internal Medicine in

late October 2026 looked at what happened when one health system made an

AI scribe available to its outpatient clinicians without restrictions.

Across a 14-hospital system, the time clinicians spent with notes open in

the EHR dropped modestly after they enrolled, most of all among frequent

users. The notes also got longer, and kept growing over the study period.

There was no change in billed work (wRVUs), and no difference in how often

notes were closed the same day or after hours.

The limitation that matters most for a signing clinician is the one the

authors state plainly: the study did not assess note content or errors.

The summary's own bottom line is that frequent use "may reduce note-active

time, although its impact on note accuracy remains unclear."

So the time spent reading the draft is not wasted time. A longer note is

more to check, and the accuracy question is still yours to answer, one note

at a time. The study covered general outpatient care, not allergy, but an

allergy note has its own places where a draft can go wrong.

Six checks for an allergy note before you sign

1. Allergens and reactions. Does the note name the right allergens,

and the reaction the patient actually described (hives, wheeze,

swelling, anaphylaxis), with timing? A reaction history that is too

vague or too severe changes the plan.

1. Test results. Do the skin-test or lab results in the note match the

recorded results, including the positive and negative controls? Results

should come from the testing record, not from what was said in the

room.

1. Immunotherapy details. For a shot visit or a build-up change:

extract, vial, dose and any reaction since the last injection. These are

the lines a draft is most likely to summarize loosely.

1. Pertinent negatives. Did the patient really deny what the note says

they denied? A negative that wasn't asked about doesn't belong in the

note.

1. Codes on each assessment item. If the draft suggests ICD-10 codes,

check each one against the assessment it sits on. A suggested code is a

suggestion.

1. Plan and handout. Does the plan say what the patient was told, and

does the take-home handout match the signed plan, medicine by medicine?

Keep the note reviewable

Length is something a practice can manage. A few habits help:

• Pick a note style and stick to it. A shorter default style for

routine follow-ups leaves less to read and less to drift.

• Don't repeat the last visit. On a follow-up, carry forward a short

summary of prior care and spend the words on what changed.

• Check against the source. The fastest way to resolve a doubtful line

is to see where it came from in the visit, not to reread the draft.

• Treat "unsure" as a feature. Software that shows the doctor what it

is unsure of, before the note or handout goes out, is easier to sign

safely than software that sounds certain.

What to ask of software that drafts allergy notes

• Can I see where a line in the note came from in the visit?
• Can each provider choose how long the note is, and does it remember?
• Are test results taken from the testing record rather than retyped?
• Are suggested codes shown on each assessment item for me to review?
• Does anything get filed without my signature?

Where Medora fits

Medora is built only for allergy and immunology practices. It drafts an

allergy-structured SOAP note from the visit, with pertinent negatives, and

each assessment item carries its ICD-10 code, ready for the doctor to

review.

• Evidence Mapping links lines of the note back to the moment in the

visit conversation where they were said, including lines in review of

systems and in past, family and social history. The clinician can click a

linked line and see its source before signing.

• Note style: each provider chooses Concise, Balanced or Detailed, and

Medora remembers the choice as their default.

• Follow-ups open with Prior care, a short summary of the last visit in

which every sentence links to the visit it came from.

• Skin-test results the nurses recorded sit beside the drafted note in

the same record, so nobody re-types them.

The clinician reviews, edits and signs every note; nothing is filed without

the clinician.

Frequently asked questions

Do AI-drafted notes save time?

One large health-system study published in October 2026 found a modest drop

in time spent with notes open after clinicians started using an AI scribe,

and notes that grew longer over time. It did not assess note accuracy, so

review before signing still matters.

What should an allergist check before signing an AI-drafted note?

Allergens and reactions, test results and controls, immunotherapy details,

pertinent negatives, the code on each assessment item, and whether the plan

and handout match what the patient was told.

How do I check a line I'm not sure about?

Go to the source. In Medora, a citation marker on a linked line shows where

in the visit conversation it was said, so the clinician can check it before

signing.

Can I make AI-drafted notes shorter?

In Medora, each provider chooses a note style (Concise, Balanced or

Detailed), and follow-ups open in the Concise style unless the provider has

chosen Detailed.

See it on your own visits

• See how Medora handles allergy documentation
• Start the 14-day free trial of Medora Scribe (no card needed)
• Book a demo
• Read the buyer's guide to AI for allergy and asthma practices
• Read the case study from a working allergy practice

*Sources: Booth SE, Gould R, Cibotti N, et al., "Impact of Unrestricted

Artificial Intelligence Scribe Deployment in a Single Health System on

Clinician Workload and Productivity: A Quality Improvement Study," *Annals

of Internal Medicine*, published online 2026-10-26,

doi:10.7326/annals-25-05294, as summarized by 2 Minute Medicine

(2026-11-05),

https://2minutemedicine.com/ai-scribe-use-associated-with-reduced-documentation-time-but-longer-clinical-notes

(read 2026-11-06). This article is general information, not clinical

advice.*