SOAP records: how to write one so that it is actually useful

The rule of separation is simple and almost always broken: the subjective holds what was reported, the objective what the clinician measured or observed, the assessment the reasoning, the plan the actions. Every time content slides into the wrong section the record loses its ability to distinguish data from interpretation, and that distinction is exactly what rereading requires.

A second principle removes half the problems on its own: every line must survive the question of who says so. The owner, the colleague from the previous shift, the instrument, the examining clinician. A record where the source of each statement can be reconstructed stays usable even when its author is unreachable.

A useful subjective carries four coordinates: what changed, since when, on what trajectory and what has already been done. Without the second and third the rest is worth little, because vomiting that started yesterday and vomiting present for three weeks with increasing episodes open different hypothesis lists. It is worth quoting the owner directly when the sentence is dense with information, because a paraphrase thins it out.

• Onset and course: date or interval of appearance, frequency, whether it is worsening, improving or stable.
• Context: current diet, outdoor access, other animals in the household, recent events such as travel or a diet change.
• Treatments already given, including those you did not prescribe, with the response observed.
• Who is reporting: the person living with the animal, a relative bringing it in today, staff from a boarding facility.
• What worries the owner, stated explicitly: it often differs from the clinical reason for the visit.

The objective is the part that ages best, provided it holds numbers rather than adjectives. Within normal limits is not a finding: it is a conclusion. At the next check, perhaps carried out by a different colleague, a value allows comparison while an adjective forces a fresh start. The same applies to absences: noting that a finding was looked for and not found is information, and separates not present from not assessed.

• Weight in kilograms to one decimal place, measured rather than estimated, on the same scale where possible.
• Body condition score on a stated scale, for example from 1 to 9, and muscle condition score, which in animals losing mass changes independently of weight.
• Vital parameters with units: heart and respiratory rate, temperature, capillary refill time, which in a well perfused animal stays under two seconds.
• Findings looked for and not found, written as such, for example no pain on deep abdominal palpation.
• Photographic documentation for skin lesions and wounds, with a scale reference in the frame.

Validated scales have an advantage no free description can offer: they produce a number comparable between operators. This holds for body condition, pain assessment, lameness and quality of life. The score must always carry the name of the scale used, otherwise a three means nothing.

The assessment has two jobs: to list the active problems and to make the reasoning explicit. The first job is mechanical and underrated. A maintained problem list, each item with an opening and a closing date, is the tool that stops the lameness being forgotten while renal insufficiency is being managed.

1. List active problems in order of clinical priority, each with the date it was opened.
2. For the leading problem write two to four hypotheses, ordered by estimated likelihood, with the reasoning that supports them.
3. State what would tell one hypothesis from another: a test, an expected finding, the response to a trial of treatment.
4. Declare the level of certainty in a single word: suspected, probable, confirmed. It changes the weight of everything that follows.
5. Record the hypotheses ruled out and why, so nobody reopens them from scratch at the next visit.

The cost of this discipline is low and the effect high: an assessment written this way lets the reader understand not only what was being considered but which information would have changed the conclusion. It is also the part that protects best when a decision is challenged, because it documents reasoning rather than a verdict.

A plan is judged by a simple test: a colleague who has not seen the patient can carry it out tomorrow without phoning anyone. To pass that test it needs four components, of which the third and the fourth are the ones most often left out.

The communication component is not defensive paperwork. An owner who declines a test or chooses a cheaper route changes the clinical pathway, and that change belongs in the record as a shared decision rather than as a failing. At the next check the same proposal returns with the new picture, and the record shows it had already been discussed.

A growing number of owners turn up with data collected at home: weekly weights, resting respiratory rate, vomiting episodes with date and time, pain scores, a video of a seizure. The material is of uneven quality and potentially high value, because it covers the interval between two visits, which is exactly the period you otherwise know nothing about.

• Treat it as structured history: it stays subjective, but with a time resolution spoken history cannot offer.
• Check the method before the content: which scale, at what time of day, before or after feeding, who took the measurement.
• Use series for trends, not for single points: a resting respiratory rate counts as a trajectory over weeks.
• Ask for video when the phenomenon is episodic: a recorded seizure is worth more than any verbal description.
• Give back a format: stating what to measure and how often turns a random collection into usable data by the next visit.

The mirror risk exists and should be named: a run of numbers collected at home can feed the idea that monitoring replaces examination. The phrasing that works in the consulting room assigns roles explicitly, namely that home data measure the trajectory between checks while the check interprets the trajectory.

A record is reread on four predictable occasions: a different shift, a second opinion, an insurance claim and a challenge. All four rely on the same formal elements, which cost seconds while writing and cannot be recovered afterwards.

1. 1

   Author and time

   Every note carries its author and the moment it was written. If a note is written later, state when the event happened and when the note was recorded.

2. 2

   Tracked corrections

   An error is corrected by adding, not by overwriting. A record where earlier text disappears without trace loses evidential value as well as clinical value.

3. 3

   Templates for repetitive visits

   A template for vaccination checks, geriatric consultations and emergency presentations guarantees that mandatory fields are always there, even on the busiest day.

4. 4

   A separate discharge document

   What goes home is written for the owner, with instructions and warning signs. It is not a copy of the clinical note, and keeping them apart improves both.

On the tooling side, the Animiyo clinic workspace includes SOAP notes, clinical records, record templates, treatment plans with interaction checking, agenda and availability, a reception queue with triage, a ward whiteboard, an anaesthesia record, a discharge form, automatic recalls, laboratory reports shared with the owner, estimates and invoices, clinic statistics and an audit log for access traceability.