For all of the Way to Health veterans out there, remember the days when the development team had to custom build your study's target generators for you? Well look no further, YOU now have the ability to configure the target generator for your study!
What is the Enrollment Run-in Target Generator?
- Calculates a participant's target or goal for the study duration on data collected during the run-in period.
- This allows participants to have a realistic and attainable goal based on real data
- Allows each participant to have a different goal based on how they did during the run in
- Ability to set different goal parameters including
- Set minimum days needed to calculate goal
- Exclude certain values
- Multiply baseline calculated to increase goal for the study period
- Specify goal rounding criteria
...
- If you have not so already, first you should set up the Run-In period in enrollment. This addition is necessary in order for WTH to calculate a goal based on the baseline data captured during enrollment.
- Not sure how to do this? See the + Enrollment Block section of How to build the enrollment processOld Enrollment Process
- Not sure how to do this? See the + Enrollment Block section of How to build the enrollment processOld Enrollment Process
- Navigate to the 'Target Configurations' section of Manage Study.
- Under Target Configurations, select 'New'.
- The Target Configuration screen will pop up. In the 'Target Type' drop down, select 'Enrollment Run-in Target'.
- Once selected, WTH will display the target set up fields. Fill out fields based on the way the target should work. See below for more help on how to fill out the various sections!
Target Configuration main settings
- Device Data Source: Select the device that the target should look at. This will be the same device that is used during the enrollment period.
- Example: Run in for 2 weeks of Fitbit pedometer data? Select Fitbit
- Haven't set up your devices yet? Go to the 'Devices' section of Manage Study.
- Device Selection Data Field: Select the data that the target should look at. This should be the same selection field that is set up in the Device Settings.
- Example: Fitbit = steps, Withings Scale = weight
- Example: Fitbit = steps, Withings Scale = weight
- Arm: Choose the arm that will have this target generator. This allows you to have different criteria across arms.
- Want the same target for both Intervention and Control? Just created the same target generator twice, swapping out the arm!
- Haven't built the study arms yet? Go to 'Study Arms' section of Manage Study
Scheduling
- Start Day: Select the day that the target should start. If you only have one target, this is typically at study start, or Day 0.
- End Day: Select the day that the target should end. Again, if you only have one target for the study duration, this is typically the last day in the study.
Tip! Want to create a target ramp up? You can do this with the scheduling feature! - Example: Target runs from Day 0-29 and then the target increases on Day 30-59 and increases again Day 60-89. You can achieve this by creating 3 different targets!
Criteria
- Minimum number of days needed to calculate target: Set the number of days of data needed to calculate a participant's target
- Example: If a run-in is 2 weeks, we want to sure that we have at least 7 days of data from the participant's device in order to calculate the target.
- Why? Any less than 1 week of data may not accurately reflect their activity level.
- Why? Any less than 1 week of data may not accurately reflect their activity level.
- Example: If a run-in is 2 weeks, we want to sure that we have at least 7 days of data from the participant's device in order to calculate the target.
- Evaluate: Choose either entire run-in (i.e. all 14 days) or only the last X of the run-in (i.e. last 10 days)
- Why would I choose only last X days? This option allows participants to get used to wearing, syncing, and charging the device. It allows them a few days to get 'acclimated' to having a new device, scrapping the earlier data that might not reflect their true activity levels.
Calculation
- Exclude values less than: Allows you to set the minimum value accepted to calculate the goal
- This field can also be left blank
- Example: for a step target, exclude values less than 1000. Why? Anything less than 1000 probably doesn't truly reflect the participant's normal step count, or the participant was sick, or is having device issues
- Exclude values greater than: Allows you to set the maximum value accepted to calculate the goal
- This field can also be left blank
- Example: for a step target, exclude values greater than 20,000. Why? Anything more than 20,000 may make someone's goal too high and unattainable once it's multiplied. You may want to get rid of the high values so that they don't skew participant's step goal.
- Multiply run-in target by: Allows you to set how much the value should increase by based on the average calculated from the baseline.
- This field can also be left blank - the participant's target will then just be an average of their run-in
- Example: for a step target, multiply run-in by 1.2. This will increase the step count by 20%.
Rounding
- Target value should round: Select if the target should round to the nearest value or up or down
- Target value should round to: Select the decimal place that the target should round to
- Tip! For steps, choose a whole number, i.e. ones, tens, or hundreds. For weight, choose a decimal, i.e. tenth
Example: Daily Average: 7,808.86 steps/day
Should round Should round to Result to the nearest tenth 7,808.9 Up tenth 7,808.9 Down tenth 7,808.8 to the nearest one 7,809 Up one 7,809 Down one 7,808 to the nearest ten 7,810 Up ten 7,810 Down ten 7,800 to the nearest hundred 7,800 Up hundred 7,900 Down hundred 7,800
Once you are done building your Enrollment Run-in target generator. Click 'Save'. The new target will now appear in your Target Configurations overview page.
Now that you have theTarget Generator built, you can use it in Study Arm event logic! For more on building events and adding logic, see Events by Type
