Methodology

This section contains methodological information for the study.

PRELAUNCH STATUS

TO DO:

  • Update CT.Gov office DONE
  • Set up Participant Payments
  • Recruitment plan
  • SOP for fully putting someone on study using STAR
  • final review of all implementations below

Decisions

  • Raw data storage will be at both the subid level and aggregate
    • both located on R drive
  • Processed data on S drive

Data Pipeline Overview

Data collection tools:

  1. Qualtrics forms
  2. NDA GUID tool
  3. FollowMee
  4. PRISM (by running scripts)

Staff data manipulation for running the study:

Susan: Update this section with path/input/output like on prism page

  1. create_bulk_guid.qmd: creates identity file from Qualtrics onboarding script for NDA GUID creation tool
  2. format_nda_data.qmd: Creates the formatted template files for uploading data to NDA
  3. NEED SCRIPT.R: update_participants.R AND/OR qmd. (Might need both)
  1. Updates a PRISM participant list on the R drive automatically so we have a running record of participants (PRISM list will only have on-study?)
  2. Also needs to be run by an RA manually after consent/interview to update when someone is added to or taken off study.
  1. NEED SCRIPT: check_adherence.qmd (start from Risk2 script)
  2. NEED SCRIPT: make_payment.qmd (start from Risk2 script)
  3. NEED SCRIPT: enrollment_rppr.qmd (creates updated RPPR enrollment csv)

Data analysis:

  • TBD

Qualtrics forms

Participant facing

  • Screening
    • 2026_0422 TB confirmed qualtrics implementation as definitive, other docs updated to match
    • needs lab review
    • use responseID as screen_id which can be carried forward to consent (inserted in URL by RA)
  • Consent
    • needs lab review
    • will need to be updated post-IRB change approval
    • DONE needs forwarding to onboard survey
  • Onboard
    • needs lab review
    • collects phone number and address for EMA & payment
    • may change per new participant payment regulations
    • DONE needs forwarding to birth data survey
  • Birth data
    • needs lab review
    • subid assigned here, unless decided otherwise
  • Baseline measures
    • SW final review to match 2026_0309
      • keep 2 optional questions on flourishing?
      • inclusion and placement of STAR questions on trust_id?
    • needs lab review
    • DONE add subid as embedded data
  • Mid/end repeated measures
    • 2026_0422 TB confirmed matches qmd measures except:
      • message_usefulness has a five-point scale, but message.qmd says it should have a seven-point scale - do we alter?
    • needs lab review
    • add subid as embedded data
  • Daily EMA
    • Finalized
    • Location context (part of EMA after day 7)
      • Finalized

The following are not finalized in qualtrics as significant revision is happening:

  • Daily Feedback
    • needs update when we get final models
    • modify for the survey to be rebuilt from data daily by script, rather than have qualtrics pull from PRISM

RA facing:

  • RA session form
    • collects daily EMA times
  • Date log form
  • WOG
  • TLFB (gathered by interview)
    • Needs calendar added and finalization of what is to be collected (only n drinks, also hours drinking, etc)
  • Exit Interview (gathered by interview)

FollowMee

Input:

  1. Installed with token on participant phones
  2. Administration done from web dashboard

Output:

  1. 3-14 days of data.
  2. There will be 24x daily automated data pulls by script, saved as daily and study-longaggregate csv files (subid level vs all aggregate?).

Scripts

These live in the proj_optimize repo where they will be called by PRISM. More will be added as they are completed

automation:

  • task_scheduler_opt.R - schedules R scripts to run
  • daily_download_opt.R - downloads GPS
  • mak_daily_qualtrics.R - downloads survey data from qualtrics including EMA
    • part of it will download EMA times from the RA session survey to be used by PRISM

scripts_staff:

  • create_bulk_guid.qmd - formats a dataframe downloaded from qualtrics to be used in the NDA GUID tool
  • format_nda_data.qmd - formats qualtrics data for NDA upload
    • format_nda_templates.R (helper script for above)

scripts_prism:

  • find_new_locations_opt.R - pulls user GPS data and finds new locations to query in EMA location context questions

NEED TO ADD:

  • John’s model script
  • recommendation script
  • mak_payments
  • chk_adherence

PRISM

Input:

  1. See csv files listed here - NB these can be updated from script OR manually

Output:

  1. EMA messages linked to a qualtrics survey with location questions
  2. See csv files listed here

NDA GUID tool

Input: RA batch loads CSV file pulled from Birth Data qualtrics form (Currently only Susan can do this) Output: CSV file with GUID added for use in NDA

OnCore

Input:

  • Participant ID information (batch upload or manual?)
  • Participant eligibility information
  • Participant visit information

Output: Whatever the CT oversight group needs - Not for study data collection, although it will provide visit information

SOPS

See SOPs

Recruitment

Soft Launch

  • Craiglist ads in Madison/Milwaukee/Chicago
  • Possibly bulk email to UW staff & students?

Advertising

  • Craigslist
  • Reddit
  • Clinic partnerships, if we can find them