Patient-completed lifestyle questionnaire: existing archetypes?

Hi everyone,

At the University Hospital Basel we are developing a patient-reported comprehensive anamnesis questionnaire, which patients complete before their outpatient consultation. Besides the usual medical items, it covers:

  • Physical activity: time spent in sedentary activities (hours per day), leisure-time physical activity in general and at moderate and high intensity (hours per week), use of a step counter (y/n), and average daily step count
  • Nutrition: number of cups of tea/coffee per day, volume of sweetened and unsweetened drinks per day, frequency of alcohol consumption (<2x/week, 2–5x/week, >5x/week), portions of fruits and vegetables per day, frequency of meat, fish and fast food consumption (<2x/week, 2–5x/week, >5x/week), type of fat/oil mainly used for cooking (butter, olive oil, etc.), and preference for salty food (y/n)
  • Sleep, stress & mood: sleep duration (hours/night on weekdays), sleep feels restful (y/n), perceived high stress in private or professional life (y/n), and two depression screening questions (during the past month: often feeling down, sad, depressed or hopeless (y/n); noticeably less interest or pleasure in things usually enjoyed (y/n) (Whooley et al., 1997))
  • Digital health: use of digital health apps and their purpose (medication adherence support, blood pressure measurement, sleep analysis, heart rate/rhythm analysis, activity tracking)

We are using the screening questionnaire family of archetypes to model the whole questionnaire, however we would need additional topic-specific screening archetypes to cover the topics/questions above. For physical activity, we are aware of openEHR-EHR-OBSERVATION.physical_activity_screening.v1, but we could not fit the questions from our use case into it.

Our questions from our side:

  • Has any work been done on a nutrition screening questionnaire, or on screening archetypes for sleep, stress or digital health use?
  • How would you recommend handling questions that don’t fit the existing physical activity screening archetype: extending it, or creating a new one?
  • Is anyone working on something similar? We’d be happy to share our work and collaborate.

Thanks in advance!

@Olha_Nikolaieva and @maren.diepenbruck

Hi Maren, welcome to the community! :slightly_smiling_face:

I’m not aware of any modelling of screening questionnaire archetypes for those specific topics, but @siljelb @varntzen or @heather.leslie might know better.

Re Physical activity screening, what would you say are the main misalignments between your requirements and the existing archetype?

Hi Maren,

The screening questionnaires are largely focused on very common and basic questions that essentially require a yes/no answer. You’ll see essentially that same repeating pattern underlying all of them.

They are useful at a very basic level for screening purposes, but as soon as they require more specific detail we rely on more formal documentation using ENTRY archetypes. For example, shortness of breath can be identified as present and the Symptom archetype should be used to record any further detail.

As far as more specific details about physical activity is concerned, I would suggest you focus your efforts on developing the OBSERVATION.physical_activity archetype further, rather than enhancing the screening questionnaire. It is rather sparse at the moment, and I have to admit I found it hard to identify how to model it. Questions that are asked of patients are usually so variable and the event model in the OBSERVATION will help. While time spent in sedentary activities per day may be common, they are often anchored by different time frames - in the last week, month, six months etc. This may be best modelled flexibly using interval events in a template and leaving the questions agnostic of any time limitations.

We also need to be mindful of the data people will be downloading from their devices. I know that @erik.sundvall was guiding some work on this some years ago. Finding a pattern to record appropriate ways to record all of the activities and the appropriate way to record distance, weight, laps, repetitions etc was not simple. See Physical activity archetypes; exercise, steps etc from apps & devices

Depression - there are a couple of specific depression scores in CKM - Edinburgh and Kutcher.

I think @Silje was doing some investigation into modelling sleep. And there are two sleep-related proposals - Proposal for Observation Archetype: Pittsburgh Sleep Quality Index (PSQI) [openEHR Clinical Knowledge Manager] and Proposal for Observation Archetype: Epworth Sleepiness Scale (ESS) [openEHR Clinical Knowledge Manager]

Nutrition - we have a Food and nutrition summary which is describing some persistent data but, again, it sounds like you need a Nutrition OBSERVATION to record the actual data on a point in time or interval basis. I don’t know of any work happening in this area yet.

Hope this helps

Heather