What Goes in the Record: Drafting Therapy Notes That Survive an Access Request

A therapist finishes a session and opens the note screen. It is one box. Into that box go the things the client could reasonably read tomorrow, the things that are the therapist’s own thinking rather than a clinical fact, and, because the session involved a marriage, several things about a person who is not a client and never agreed to be in anyone’s file. The software offers no way to distinguish them, so the therapist makes the distinction in their head, differently on a tired Thursday than on a fresh Monday.

This is the transformation we would run for that practice. It is presented as a concept, built on the same method we use in client engagements, so the before, the after, and the path between them are all visible.

Why is the hard part of a therapy note not the writing?

Most documentation tools for clinicians solve dictation. In counselling, dictation is the easy half. The difficult half is placement, because a counselling record has readers a physical examination note does not.

In Ontario an individual generally has a right of access to their own record of personal health information, subject to limited exceptions. That single fact changes what a note is. It is not a private aide-memoire; it is a document with a foreseeable second reader who is the person it is about. A counselling note also routinely contains information about someone else entirely: a partner, a parent, a colleague described in detail during a session, none of whom consented to being recorded anywhere.

Practices handle this today through therapist discretion, which works exactly as well as discretion under caseload pressure ever does. The frequently repeated statistics about how many hours therapists lose to documentation all trace back to companies selling documentation software, so this study does not quote them. The direction is not in doubt; the number is not sourceable, and inventing one would be the same error the rest of these pages refuse to make.

Before and after: the note

Drag the handle. The before is the single free-text box; the after is the note designed with the boundary visible.

The therapist writes no less and thinks no less. What changes is that placement becomes an explicit, reviewable decision instead of an invisible one made at speed.

What we built

The concept above is the output of the same engagement steps we run in client work, and several are decisions a competent peer could argue with.

• We split one field into three, and made the split a first-class part of the interface rather than a setting. Shared clinical record, therapist working note, and content naming an identifiable third party are different objects with different readers. A practice can disagree with where a given line goes; it can no longer fail to notice that the question exists.
• We have the system propose placement and refuse to commit it. Every proposal shows what triggered it, and the therapist accepts or moves it. Nothing is filed, and crucially nothing is removed, without a person.
• We encode the practice’s own rule rather than a general default, and where the practice has no rule the system flags instead of guessing. Discovering mid-engagement that a group practice has no written rule for couples work is a real and uncomfortable outcome, and it is better than discovering it during an access request.
• We deliberately left the therapist working note unstructured and unsummarised. It is the one place a clinician thinks in their own words, and putting a model inside it would change how people write there. That is the decision most likely to be challenged as a missed opportunity, and we would make it again.
• We wire evals from day one against the practice’s own historical notes, scored on whether proposed placement matched what an experienced therapist chose, and every override becomes a test case.

How it works

The practice management system stays the source of truth for the clinical record. After a session, the draft is produced and each passage is proposed into one of three destinations with the reason attached: this names an identifiable third party, this reads as clinical formulation rather than fact, this is a clinical observation. The therapist confirms or moves each one. Only then is anything filed, and the working note remains outside the shared record by construction rather than by convention.

Because session content stays in Canadian infrastructure and processing runs with zero data retention, the design lines up with what PHIPA expects: a privacy impact assessment before deployment, an agreement naming the processor as an agent under section 17, and an audit log on every access. Retention is set short.

Why this pays back

The operational return is the hour a therapist stops spending re-reading their own notes to work out what they can safely release, and the clinical return is that the boundary is consistent on a Thursday. Break-even for the first slice is about twelve clinical hours returned a month at the practice’s own billable rate, which the practice supplies.

The limit worth saying before anyone signs, and it is capable of losing the sale: this makes access requests survivable, not comfortable. A practice whose real concern is that its historical records are a mess will not be helped much, because nothing here retroactively fixes six years of single-box notes. That is a records project, it is expensive, and it is honest to say so before rather than after.