Utilizing the Patient-Specific Needs Evaluation for Data-Supported and Individualized Treatment (UNITE)

Author(s)

Robbert Wouters, PhD1, Willemijn De Ridder, phd2, Nina Loos, MSc2, Yara Van Kooij, Msc2, Lisa Hoogendam, MSc3, Harm Slijper, PhD4, Ruud Selles, PhD2.
1Assistant professor, Erasmus MC, Rotterdam, Netherlands, 2Erasmus MC, Rotterdam, Netherlands, 3Erasmus Medical Center, Hellevoetsluis, Netherlands, 4Equipe Zorgbedrijven, Delft, Netherlands.
OBJECTIVES: Clinical decisions often rely on clinician preferences and guidelines. This can lead to subjective and one-size-fits-all decision-making insufficiently aligned with individual patient goals and needs. We introduce the Patient-Specific Needs Evaluation (PSN) as a new framework for data-supported, patient-centered, value-based healthcare. The PSN is a patient-specific but generic tool that identifies: 1) the patient’s most important information needs, 2) treatment goals, 3) the individual’s threshold for being satisfied with treatment results (Personal Meaningful Gain, PMG), and 4) whether this is achieved at follow-up. This abstract outlines the development, validation, and application of the PSN in predictive modeling for hand/wrist care
METHODS: Following a user-centered mixed-methods approach and COSMIN guidelines, we developed and validated the PSN via pilot surveys (prototype: n=223, final: n=275), cognitive debriefing (n=16), and validation (n=2087). We assessed whether the PMG was more predictive of being satisfied with treatment results compared to the Minimal Important Change (MIC) and Patient Acceptable Symptom State (PASS, n=5373). We also developed an individualized decision-support model predicting the probability of obtaining the PMG using various machine learning algorithms.
RESULTS: The 5-item PSN (±3 minutes), is valid, reliable, and accessible online (https://personeel.equipezorgbedrijven.nl/ls/index.php?r=survey/index&sid=587344&lang=en). The PMG had a higher positive predictive value (0.76, [0.74-0.78]) for being satisfied compared to the PASS and MIC for pain (PASS: 0.70 [0.69-0.72], MIC: 0.64 [0.62-0.65],p<0.001) and function (PASS: 0.67 [0.65-0.68], MIC: 0.64 [0.62-0.66],p<0.001). We selected a gbm algorithm with 13 predictors (AUC: 0.73[0.70-0.76], excellent calibration) and developed a Shiny app to visualize predictions (https://veni-modellen.timformatie.nl/).
CONCLUSIONS: We present the PSN as a generic, yet patient-specific instrument to tailor care to the patient’s individual needs and goals, while also evaluating these needs and goals. This offers a new paradigm for shared decision-making and patient-centered care, with potential for predictive modeling and expansion beyond hand and wrist care, which is currently underway.

Conference/Value in Health Info

2025-11, ISPOR Europe 2025, Glasgow, Scotland

Value in Health, Volume 28, Issue S2

Code

PCR261

Topic

Clinical Outcomes, Patient-Centered Research, Study Approaches

Topic Subcategory

Instrument Development, Validation, & Translation, Patient-reported Outcomes & Quality of Life Outcomes

Disease

Musculoskeletal Disorders (Arthritis, Bone Disorders, Osteoporosis, Other Musculoskeletal), Personalized & Precision Medicine, Surgery

Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on ispor.org. Update my browser now

×