How should we measure physical activity after stroke? An international consensus

Background: Physical activity is important for secondary stroke prevention. Currently, there is inconsistency of outcomes and tools used to measure physical activity following stroke. Aim: To establish internationally agreed recommendations to enable consistent measurement of post-stroke physical activity. Methods: Stroke survivors and carers were surveyed online once regarding what is important in physical activity measurement. Three survey rounds with expert stroke researchers and clinicians were conducted using Keeney’s Value-Focused Thinking Methodology. Survey 1 identified physical activity tools, outcomes, and measurement considerations which were ranked in Survey 2. Consensus recommendations on tools were then formulated by the consensus group based on survey responses. In Survey 3, participants reviewed ranked results and evidence gathered to determine their support for consensus recommendations. Results: Twenty-five stroke survivors, 5 carers, 18 researchers, and 17 clinicians from 16 countries participated. Time in moderate-vigorous physical activity and step count were identified as the most important outcomes to measure. Key measurement considerations included the ability to measure across frequency, intensity, duration domains in real-world settings; user-friendliness, comfort, and ability to detect changes. Consensus recommendations included using the Actigraph, Actical, and Activ8 devices for physical activity intensity; ActivPAL for duration and Step Activity Monitor for frequency; and the IPAQ and PASE questionnaires. Survey 3 indicated 100% support for device and 96% for questionnaire recommendations. Conclusions: These consensus recommendations can guide selection of physical activity measurement tools and outcomes. Tool selection will depend on measurement purpose, user-knowledge, and resources. Comprehensive measurement requires the use of devices and questionnaires.

The following questions will ask you to identify key elements to consider in post stroke physical activity measurement for research purposes related to six categories.Please identify all key elements that you think may be relevant.Please consider both devices and questionnaires or surveys in your responses.You may or may not wish to specify different answers for each.We have collated the responses from the Round 1 survey to inform this round (Round 2).This survey will ask you to rank items.
In the previous survey you were asked identify key elements to consider for post stroke physical activity measurement in clinical practice related to six categories.
We have collated the identified key elements for consideration in each of the six categories.We now ask you to rank them in order of importance for physical activity measurement in stroke research within each of the six categories.
Please note you cannot rank items equally.
For the category "Relevance of the measurement tool's intended purpose to its intended use (construct validity)" please rank the following considerations from most important ("1") to least important ("9") for physical activity measurement in stroke research.Is the data provided free from errors?
Can you please provide a rationale for the item you ranked as most important for "Reliability" for objective physical activity measurement.__________________________________________ For the category "Reliability" please rank the following considerations from most important ("1") to least important ("8") for self-reported physical activity measurement in stroke research.Is the terminology clear, easy to understand and unambiguous?Are consistent instructions provided?
Are variable levels of stroke severity and cognition likely to influence results?
Can you please provide a rationale for the item you ranked as most important for "Reliability" for self-reported physical activity measurement.__________________________________________ For the category "Feasibility" please rank the following considerations from most important ("1") to least important ("11") for objective physical activity measurement in stroke research.Please note that some of these outcomes can be measured subjectively (eg.using validated questionnaires or activity diaries), some can be measured objectively (eg.using accelerometers or behavioural mapping) and some can be measured either subjectively OR objectively.
Under each category are related items.We would like you to 1) rank the related items for each category in order of importance for measurement of PA in stroke research and then 2) rank each category in order of importance for measurement of PA in stroke research.
Please note you cannot rank items equally.
For the category "Frequency" of physical activity please rank the following outcomes from most important ("1") to least important ("8").For the category "Duration" of physical activity per day, please rank the following outcomes from most important ("1") to least important ("12").We have collated the responses from the Round 1 survey to inform this round (Round 2).This survey will ask you to rank items.
In the previous survey you were asked identify key elements to consider for post stroke physical activity measurement in clinical practice related to six categories.
We have collated the identified key elements for consideration in each of the six categories.We now ask you to rank them in order of importance for physical activity measurement in stroke clinical practice within each of the six categories.
Please note you cannot rank items equally.
For the category "Relevance of the measurement tool's intended purpose to its intended use (construct validity)" please rank the following considerations from most important ("1") to least important ("9") for physical activity measurement in stroke clinical practice.

Survey 2 Clinicians
Can you capture non-ambulatory physical activity?
Can you identify the context and setting the physical activity occurred in?
Is the measure generalisable across a wide range of patients?
Can you please provide a rationale for the item you ranked as most important for "Relevance of the measurement tool's intended purpose to its intended __________________________________________ use (construct validity)." For the category "Responsiveness & Sensitivity" please rank the following considerations from most important ("1") to least important ("5") for physical activity measurement in stroke clinical practice.Is there reliability for repeated measurements?
Is the terminology easy to understand and are the instructions clear?
Can you please provide a rationale for the item you ranked as most important for "Reliability" for self-reported physical activity measurement.__________________________________________ For the category "Feasibility" please rank the following considerations from most important ("1") to least important ("11") for objective physical activity measurement in stroke clinical practice.Can you please provide a rationale for the item you ranked as most important for "Feasibility" for objective physical activity measurement.__________________________________________ What are the key elements to consider for measuring physical activity post stroke when thinking about the measurement tool's construct validity?(Does the __________________________________________ tool measure what it is intended to measure?) Eg.Can you remotely monitor physical activity?Can you determine what types of physical activity are performed?What are the key elements to consider for measuring physical activity post stroke when thinking about responsiveness & sensitivity?(ability to detect __________________________________________ change) Eg. epoch length; can it measure change in activity levels?What are the key elements to consider for measuring physical activity post stroke when thinking about reliability?(consistent and free from errors) Eg. __________________________________________ duration of measurement; recall issues What are the key elements to consider for measuring physical activity post stroke when thinking about feasibility?Eg cost; time to complete __________________________________________ What are the key elements to consider for measuring physical activity post stroke when thinking about the ability to run statistical analyses?Eg. what scales __________________________________________ of measurement do the data represent?What are the key elements to consider for measuring physical activity post stroke when thinking about relevance to the International Classification of __________________________________________ Functioning and Disability model (ICF model)?Eg. ability to truly measure participation; does the change reflect changes in body structure?Are there any other considerations for measuring physical activity in stroke research?__________________________________________ exercise (any intensity) (min/day) Habitual weekly minutes of physical activity(min/week) Time spent sitting (min/day) Time spent lying / reclining (min/day) Time spent standing (min/day) Time spent walking (min/day) Time spent ascending / descending stairs (min/day) Time spent running (min/day) Time on feet (min/day) Can you please provide a rationale for the item you ranked as most important for "Duration of physical activity."__________________________________________ Now please rank each physical activity outcome category in order of importance to measure provide a rationale for the item you ranked as most important for "physical activity outcome category."__________________________________________ Do you have any comments to add? __________________________________ everyday commercial device that is readily available?Is the device simple & easy to use (for the patient)?Is the device simple & easy to use (for the clinician)?Is the device easy to don and doff?(eg.put on & remove; where is it worn?)How comfortable is the device to wear?Is it obtrusive?Is education and training required for the therapist to use and interpret results?Are clinicians open to using the device (potential change in practice)?Does the device have a long battery life?Does the device have GPS capabilities?Does the device have minimum wear time with good reliability?