Agreed, but the length of the exam is not the most important thing. The patient will probably be the same at the start of the forty-five minutes as at the end. More important is seeing the patient over time, on different days and at different times. The doctor also has to rely on the observations of reliable staff for the assessment. If five nurses tell you the patient talked, then they're not PVS, even if the doctor didn't see it.
Right. And if a patient is cognitive on a short video clip, they're not PVS even if there's a lot of video footage where they're not.
If someone claims they have a Rembrandt landscape, and hands me a Polaroid, I could not judge from such a photo whether it was real. But if the landscape included a McDonald's in the background, even a crummy photograph would be sufficient to declare the painting to be fake.
Likewise, video can't confirm a PVS diagnosis, but it can certainly refute one.