Many online consultation systems force patients to answer lots of multiple-choice questions (MCQs) about their symptoms. They then direct them to A&E or 999 if they give a particular answer.
Patchs is different.
We allow patients to write about their symptoms as free-text answers to open-ended questions, using their own words.
If switched on, Topic AI Questions can ask patients further questions based on the clinical topic of their request. Meanwhile, Signpost AI can signpost patients away in case of an emergency. This is done based on the general content of their request however (not on their specific answers), and patients can choose to not answer or override signposting advice.
Why do we think it's better if patients can give free-text answers?
There are several reasons:
- It's more like a traditional GP consultation. Our open-ended questions are typical of the ones that GPs and other primary care clinicians use. They include symptom descriptions but also ideas, concerns, and expectations. Additionally, free-text responses make it conversational: clinicians want to hear patients describe their problems in their own words. This is more natural than mechanical yes/no answers to MCQs.
- Patients prefer it. Patients don't like MCQs in online consultations because they're difficult to understand, long-winded, inflexible, and hard work. They can find it hard to fit themselves into a limited set of categories. Often, they are also asked 10s of questions for a simple complaint, many of which are completely irrelevant. Patients will often give up on MCQ online consultations halfway through and call the practice instead, which completely defeats its purpose.
- It's less work for GP practice staff. When patients fill out MCQs in online consultations, it increases workload for staff because: 1) there's lots of questions and answers to read (many of which are irrelevant), 2) despite asking lots of questions, MCQs often don't provide sufficient detail about the patient's request, meaning that you end up contacting them for further information anyway by phone, 3) you are responsible to further assess any irrelevant symptoms which they say they have, 4) patients can 'game the system' and answer MCQs in a particular way to get what they want (e.g. a face-to-face appointment for a condition that could be managed by self-care).
- It's safer. 1) When patients describe their problems and request in their own words they are more expansive. It provides a richer narrative than just one-word answers to MCQs. This can provide more relevant information to the patient's problem, meaning that you can make more informed clinical decisions. 2) If a patient is forced to contact A&E based on their answer to an MCQ, then they can just go back and change their answer, meaning that you end up with an underplayed picture of their symptoms. 3) Online consultation systems that use MCQs are attempting to ask every possible question for a patient's complaint so that you don't have to. This can lead you to rely on the system - though there will always be times it will get things wrong.
You can read the research evidence for the above claims our 'research evidence' section.
