PROPr
The Patient Reported Outcomes Measurement Information System (PROMIS) represents a major improvement in descriptive health-related quality of life measurement. This system has developed measures using Item Response Theory (IRT). IRT estimates a continuous latent construct, such as physical functioning.
This project is the first to place a preference-based function on the continuous latent constructs as measured in PROMIS. The PROMIS-Preference (PROPr) score is a generic, societal, preference-based summary score. The PROPr score can be estimated if an analyst has scores for the following PROMIS domains:
- Cognitive Function – Abilities Subset (v2.0 or Cognitive Function v2.0)
- Emotional Distress – Depression (v1.0 only)
- Fatigue (v1.0 only)
- Pain Interference (v1.0 or v1.1)
- Physical Function (v1.0, v1.1, v1.2, or v2.0)
- Ability to Participate in Social Roles and Activities (v2.0 only)
- Sleep Disturbance (v1.0 only)
There are three recommended ways to collect the necessary data:
1. Use computer adaptive testing to administer each domain.
2. Use the PROMIS-29+2 Profile v2.1 (PROPr). This is the PROMIS-29 Profile v2.1 and two items from Cognitive Function – Abilities v2.0
3. Use the PROMIS-29 Profile v2.1 and a PROMIS Cognitive Function – Abilities or PROMIS Cognitive Function short form (e.g., 4a).
The Technical Report v1.4 for this project, including scoring code for R and SAS, is available here. R and SAS code are also available on Github.
Key peer-reviewed references for this project include:
The PROMIS of QALYs
Selection of key health domains from PROMIS® for a generic preference-based scoring system
Evaluation of options for presenting health-states from PROMIS® item banks for valuation exercises
Estimation of a Preference-Based Summary Score for the Patient-Reported Outcomes Measurement Information System: The PROMIS®-Preference (PROPr) Scoring System
Cross-sectional validation of the PROMIS-Preference scoring system.
Some data associated with this project are available on the Open Science Framework.