There is a decent argument that the single most under-used piece of equipment in physiotherapy is the second measurement. The first one gets taken carefully, at assessment, with the goniometer out of the drawer. The second one, six sessions later, gets estimated by eye, because finding the first one would mean scrolling through notes.
Everything below is about closing that gap, and almost none of it is about buying a better measuring tool.
The measurement matters more than the tool
A plastic goniometer, a phone app and an expensive digital inclinometer will all give you a number for shoulder flexion. They will disagree by a few degrees. That disagreement matters much less than it feels like it should, for a simple reason: you are not comparing your number to a textbook. You are comparing it to your own number from three weeks ago.
Which means the property you actually need is repeatability, not accuracy. A tool that reads three degrees high every single time is more useful than one that is perfectly calibrated and used differently on Tuesdays. And repeatability is mostly about how you take the measurement, not what you take it with.
Phone goniometers: how accurate, really
Modern phones have good inclinometers, and for movements measured against gravity — shoulder flexion and abduction, hip flexion, knee flexion in lying, trunk side-flexion — a phone app held against the segment is comparable to a hand-held goniometer for practical purposes. Published comparisons generally put the difference in the range that clinicians already accept between two of their own measurements.
Where phone apps get worse, and it is worth knowing which:
- Rotation in the horizontal plane. Gravity cannot tell you anything about it, so the app is either using the compass or asking you to hold the phone in a way that introduces its own error.
- Small joints. A phone against a finger is a phone against three joints at once.
- Anything needing two hands. Stabilising the scapula with one hand and holding a phone with the other is a compromise, and it is always the stabilising that loses.
The practical answer most clinics land on: phone for the gravity-dependent large joints, a physical goniometer for the rest, and consistency about which one was used — because switching tools between sessions manufactures a change that did not happen.
Pain scales, and the trap of one number
A single 0–10 score is the most recorded and least informative measurement in the profession. “Six” on its own tells you nothing you can treat. Three small additions make it useful, and none of them takes extra time:
- Say when. Pain now, pain at worst this week, and pain with the movement you are testing are three different numbers, and only the third responds to what you did today.
- Say what it stops. “Six, cannot sleep on that side” is a measurement you can retest. “Six” is a mood.
- Record the zero honestly. A blank is not a zero. If you did not ask, leave it empty — a record full of default zeros produces a graph that looks like a recovery nobody had.
Put one real course of treatment in
An initial assessment, four progress notes and a retest. Then look at whether the change is visible without you doing arithmetic.
Create a free account → No credit card. The first 7 days have no limits at all.Which outcome measures earn their two minutes
The honest position is that most private practices do not need a battery of standardised questionnaires. They need two or three things measured consistently. A reasonable minimum, per patient:
- One objective measurement relevant to the problem: a range, a strength grade, a distance, a time.
- One functional test the patient cares about: sit-to-stand in thirty seconds, a single-leg heel raise count, how far before the back gives out.
- One patient-reported item in their own words: what they still cannot do. Rewritten at reassessment, unchanged wording, so the difference is readable.
Region-specific questionnaires — the usual shoulder, knee and back indices — are genuinely better evidence and worth using if you will actually score them every time. A questionnaire administered once at assessment and never repeated is paperwork, not measurement.
A measurement that never becomes a curve
This is where most measurement effort is lost, and it is a records problem rather than a clinical one. If the number lives in the prose of a note — “shoulder flexion 120 degrees today” — then comparing session one to session eight means reading eight notes. Nobody does that during a treatment session. So the comparison does not happen, and the measurement might as well not have been taken.
What makes it work is dull: the same measurement in the same field every time, so the software can line them up. Then the question “is this person better than a month ago” is answered by looking, not by remembering.
If you are choosing software, this is worth testing directly. Enter four sessions of the same measurement and see whether the product shows you a trend or just four notes containing a number.
Showing the patient their own graph
There is a clinical use for this beyond your own reasoning. Patients in week five of a twelve-week problem generally cannot feel the improvement — the pain is still there, the thing they want to do is still not possible, and the honest summary of their experience is “this is not working”.
Showing them a line that has moved from 95 degrees to 130 changes that conversation completely. It is the same information they already had, made visible, and it is one of the cheapest interventions available against people quietly stopping treatment halfway. Practices that show patients their own numbers report fewer of the dropouts that nobody announces.
Same position, same time, same person
Four habits do more for your measurements than any app:
- Write down the position, once, in the record, and reuse it. Supine versus standing is a bigger difference than any tool error.
- Measure at the same point in the session. Range after twenty minutes of treatment is not range on arrival, and comparing one to the other invents progress.
- Same person where possible in a multi-clinician practice, or an agreed written protocol where it is not.
- Retest on a schedule, not when you happen to remember. Every fourth session, or at a fixed review point, decided in advance.
What Physioly records, honestly
Physioly holds the measurement side of this: assessment forms by specialty, a progress note per session, a pain score that is charted across the course of treatment so the change is visible without arithmetic, a body map, functional test fields and file attachments for photographs or reports. The pain score starts empty rather than at zero, which sounds like a detail until you have seen a graph built out of default zeros.
What is not there:
- No built-in goniometer or inclinometer. You measure with whatever you use and type the number in.
- No import from measurement apps. There is no connection to third-party goniometer tools.
- No automatic scoring of standardised questionnaires. You can record the score; the product does not calculate it for you.
- No video movement analysis.
So the honest summary: it is the place the numbers live and turn into something you can look at, not the instrument that takes them. If what you are shopping for is the measuring tool itself, a dedicated app will beat this and cost nothing. If the problem is that your measurements are scattered across notes and never compared, that is the part this fixes, and the free plan costs nothing forever with an unlimited first week.
On what to collect at the first appointment rather than how to remeasure it, the intake and assessment article covers the other half.
Questions we get asked
Are phone goniometer apps accurate enough for clinical use?
For movements measured against gravity — shoulder flexion and abduction, hip and knee flexion, trunk side-flexion — a phone inclinometer is comparable to a hand-held goniometer in practice. They are weaker for rotation in the horizontal plane, for small joints, and whenever you need your other hand to stabilise. What matters more than the tool is using the same one, in the same position, every time.
What should a physiotherapist measure at every reassessment?
Three things are usually enough: one objective measurement relevant to the problem, one functional test the patient cares about, and one patient-reported item in their own words about what they still cannot do. Measured the same way each time, those three show change more reliably than a longer battery collected inconsistently.
Why record a pain score if the patient just says a number?
Because one number on its own is not treatable. Pain now, pain at worst this week and pain during the movement you are testing are three different measurements, and only the last responds to today’s session. Adding what the pain stops the patient doing turns the score into something you can retest rather than a mood.
How do you show a patient that they are improving?
Show them the numbers over time. Patients in the middle of a long problem usually cannot feel the improvement, and their honest summary is that it is not working. A line moving from 95 degrees to 130 is the same information they already had, made visible, and it is one of the cheapest things available against people quietly abandoning treatment halfway.
Does Physioly include a goniometer?
No. Physioly is where the measurements live, not the instrument that takes them: assessment forms by specialty, a progress note per session, a pain score charted across the whole course of treatment, a body map and functional test fields. You measure with whatever tool you use and type the number in. There is no built-in goniometer, no import from measurement apps and no automatic scoring of standardised questionnaires.