Stephen Doogan1, John Heid2, Samir Benosman3, Alexis Ogdie4, Layne Martin5, Prashanth Sunkureddi6 and Jacqueline Palmer7, 1Real Life Sciences, New York, NY, 2Kinapse, Inc, London, United Kingdom, 3Advisory Services, Kinapse, Inc, London, United Kingdom, 4University of Pennsylvania, Philadelphia, PA, 5CreakyJoints, NY, 6Rheumatology, Clear Lake Rheumatology Center, Nassau Bay, TX, 7Novartis Pharmaceuticals Corporation, East Hanover, NJ
Meeting: 2016 ACR/ARHP Annual Meeting
Date of first publication: September 28, 2016
Background/Purpose: Psoriatic Arthritis (PsA) is a debilitating disease with significant impact on patient quality of life (QoL). While patient reported outcome (PRO) instruments broadly capture patient experiences with PsA, they may not identify the full extent of issues patients face. In this study, we compared the types of experiences patients self-report online to concepts captured within commonly used PRO instruments in PsA.
Methods: We collected unguided patient narratives reported between January 2010 and May 2016 from 31 online sources. 56% of narratives were from health social networking sites, 22% from disease-specific patient forums, 9.9% from general health forums, and 0.1% from mainstream social media. Using natural language processing combined with manual curation, we evaluated patient reported experiences within narratives and categorized them into 5 high level themes: Social (e.g. family burden), Physical (e.g. pain), Emotional (e.g. anxiety), Cognitive (e.g. memory), and Role Activity (e.g. work), or SPEC-R. For comparison, the same SPEC-R categorization was applied to 3 PRO instruments: the Psoriatic Arthritis Quality of Life tool (PsAQoL), Psoriatic Arthritis Impact of Disease tool (PSAID), and Short Form 36 Health Survey (SF-36). The instruments were then evaluated based on their capacity to capture the concepts and sub-concepts extracted from patient narratives.
Results: 15,390 narratives from 3,139 patients were qualified into the sample for analysis. Across the SPEC-R categories, Physical concepts were reported by 81.6% of patients, Emotional concepts by 60.7%, Cognitive concepts by 20.0%, Role Activity concepts by 8.1%, and Social concepts by 5.6%. In addition, 66.8% of patients reported general concepts (e.g. “Feeling unwell”) that could not otherwise be classified into the SPEC-R categories. Pain and dermatology related issues were the most reported physical sub-concepts (66% and 26%, respectively). Emotional concepts were mainly comprised of anxiety (60%) and depression (38%). Cognitive concepts were characterized by impulsive behavior (25%) and balance/coordination issues (12%). Patients reporting Role Activity concepts focused on work/school performance (55%) and parenting (17%). Among patients reporting Social concepts, family burden (50%) and relationships with spouses (27%) were most frequent. While PRO instruments capture many physical and social concepts and sub-concepts, many sub-concepts within other SPEC-R categories were not captured by either PsA-specific or general health related PRO instruments (see table).
Conclusion: This study offers insight into a variety of patient experiences with PsA. The physical and emotional impact of PsA dominates self-reported patient experiences. PROs capture many of the physical and emotional impairments of PsA, but may miss key cognitive and role activity issues that affect patient QoL.
Table 1. Fit/Gap Analysis of PsA PRO Tools with regards to Reported SPEC-R Concepts and Sub-concepts
