Shadow Health Questions to Ask
Short, single-topic questions for a Shadow Health digital patient interview, in the order an assessment usually follows: chief complaint, symptom analysis, medications, allergies, past history and social history. Includes notes on phrasing the simulation tends to recognise.
The questions
Open any question for the note
What brings you in today?
Why ask it
Type the opener exactly this plainly. A long compound greeting usually returns one generic line, while the short version gets the complaint in the patient's own words, which is the phrasing your subjective note should use.
When did it start?
Why ask it
Onset is the first item the documentation asks for, and these patients give a specific answer, a number of days or weeks, only when the question is about start time alone rather than start time and progression together.
Where is it?
Why ask it
Location has to be asked separately from quality even when that feels repetitive. The answer often adds something you cannot get otherwise, such as the sensation spreading to a second area.
What does it feel like?
Why ask it
This is the quality question, and the patient's own adjective is what belongs in the chart. Do not tidy it up: writing sharp when the patient said crampy is exactly the kind of substitution that gets marked.
How bad is it, from zero to ten?
Why ask it
The number gives you a value you can chart and compare later. Ask again toward the end of the encounter if anything has changed, since a second reading is more useful than one in isolation.
Is it constant, or does it come and go?
Why ask it
Timing changes the whole picture. If the answer is intermittent, follow up with how long each episode lasts, because duration of an episode is scored separately from the pattern.
What makes it worse?
Why ask it
Aggravating factors are recorded separately from relieving ones, so ask them as two questions. Combining them is one of the most common reasons a transcript comes back with a gap in the symptom analysis.
What makes it better?
Why ask it
The answer here often uncovers self-treatment the patient never classified as medication, such as an over-the-counter remedy, a heat pack, or a relative's inhaler.
Have you had this before?
Why ask it
A first episode and a recurring one need different follow-up, so this answer reshapes the rest of your interview. If it is yes, ask what happened last time and what was done for it.
Have you taken anything for it?
Why ask it
This catches remedies patients do not count as medicine. Painkillers, herbal preparations and leftover prescriptions all belong in the medication list, and none of them reliably appear in answers to a general medication question.
What medications do you take?
Why ask it
Keep it this short. Asking about medications, doses and reasons in one sentence usually returns only the drug names, and the missing pieces count against your documentation rather than against the phrasing.
How much of that do you take each time?
Why ask it
Dose is scored separately from the drug name and appears only when asked for directly. Ask it drug by drug, then ask how often they take it as its own question, because a combined dose-and-frequency question usually comes back with only the amount.
Do you have any allergies?
Why ask it
Allergies come in a fixed sequence: whether there are any, then to what, then what happens. The plain version is what opens that sequence, so ask it before any of the detail.
What kind of reaction do you get?
Why ask it
The reaction matters more than the substance, because it separates an intolerance from a true allergy. An upset stomach and facial swelling are two very different entries in the record.
Have you ever been in hospital or had surgery?
Why ask it
Admissions and operations usually sit in a different section of the record from illnesses, so they need asking explicitly. Follow up with when it was and what it was for.
Do any illnesses run in your family?
Why ask it
Family history is scored condition by condition, so one broad question rarely fills it. After the general answer, name the common ones, heart disease, diabetes, high blood pressure, cancer, asthma, one at a time.
Do you smoke?
Why ask it
Ask plainly, then quantify. A yes needs how much and for how long; a no still needs a follow-up about past use, because former smoking belongs in the record and will not be volunteered.
How much alcohol do you drink in a week?
Why ask it
Quantifying in the question avoids a yes or no that documents nothing. The same phrasing works for other substance use, and it is easier for the patient to answer than a question about whether they drink at all.
Who do you live with?
Why ask it
Living situation belongs in the social history and also shapes the plan you write. Who is at home determines whether anyone can help with medication, transport or follow-up appointments.
Is there anything else you want to tell me?
Why ask it
The closing question, and it is usually scored. These patients often hold one relevant detail back until asked openly, and it tends to be something that changes your priorities.
Interviewing a Digital Patient Efficiently
Practical guidance for the conversation itself
How the Interview Responds to Your Phrasing
One topic per question
Compound questions are the main cause of thin transcripts. "When did it start and what makes it worse" typically answers half of itself, and the unanswered half is recorded as never asked.
Shorter beats more detailed
If a question returns a vague or non-committal reply, rephrase it with simpler words rather than adding qualifiers. Extra clauses make a question harder to match, not clearer.
Nothing is inferred on your behalf
The patient answers only what you ask. Anything you assumed from the case description, including obvious follow-ups, counts as missing unless it appears in the transcript.
Empathy is usually its own category
Acknowledging what the patient has said is often scored separately from the questions. A short response after a difficult answer costs one line and is easy to forget while working down a list.
Areas the Symptom Questions Do Not Cover
- Diet and appetite: "What did you eat yesterday?" gets more usable detail than asking whether they eat well.
- Sleep: "How many hours do you sleep?" then "Do you wake up during the night?" as a separate question.
- Mood: "How has your mood been?" and, where the case warrants it, a direct question about low mood or anxiety.
- Work and exposures: "What do you do for work?" then "Are you around dust, chemicals or fumes?"
- Travel: "Have you travelled recently?" with a follow-up asking where.
- Preventive care: last check-up, immunisations, and for relevant cases, menstrual and reproductive history, each asked in its own short question.
Writing It Up Afterwards
Quote rather than translate
Keep the patient's own words for the complaint and for the quality of the symptom. Converting them into textbook terms loses the detail the note exists to preserve.
Record the negatives
Denies fever, denies shortness of breath, no known drug allergies. A negative you asked about is data; an absence in the note reads as a question never asked.
Keep numbers as numbers
Days since onset, severity out of ten, doses in milligrams, drinks per week. Vague quantifiers are the easiest thing to lose marks on and the easiest to avoid.
Separate what you were told from what you observed
Subjective is what the patient reported, objective is what you measured or saw. Mixing them is a common error even when the interview itself went well.
Common Mistakes
Leading questions
"You don't smoke, do you?" invites the answer you signalled and records almost nothing. Ask neutrally and let the answer be whatever it is.
Clinical vocabulary
Asking about dyspnoea, syncope or nocturia usually gets confusion rather than information. Use the words a patient would use, then chart the clinical term yourself.
Repeating questions you already asked
The transcript shows every attempt, so duplicates read as disorganisation. Keep a rough checklist beside you rather than relying on memory mid-interview.
Following an interesting thread too far
It is easy to spend most of the encounter on one symptom and never reach allergies or social history. Finish the symptom analysis, then move on deliberately.