P548 External validation and consistency in time of patient segmentation based on disease acceptance and perceived control in Inflammatory Bowel Disease
Van Erp, L.W.(1);Thomas, P.W.A.(2);Groenen, M.J.M.(1);Bloem, S.(3);Russel, M.G.V.M.(4);Römkens, T.E.H.(5);Wahab, P.J.(1);
(1)Rijnstate Hospital, Crohn & Colitis Centre- Department of Gastroenterology and Hepatology, Arnhem, The Netherlands;(2)Radboud University Medical Centre, Department of Gastroenterology and Hepatology, Nijmegen, The Netherlands;(3)Nyenrode Business University, Center for Marketing & Supply Chain Management, Breukelen, The Netherlands;(4)Medical Spectrum Twente, Department of Gastroenterology and Hepatology, Enschede, The Netherlands;(5)Jeroen Bosch Hospital, Department of Gastroenterology and Hepatology, 's-Hertogenbosch, The Netherlands;
Background
The patient segmentation model based on disease acceptance and perceived control may guide personalized inflammatory bowel disease (IBD) care. A first single-centre evaluation of the model showed its validity and found that disease acceptance and perceived control are important determinants of health-related quality (HRQoL) in IBD patients. We aimed to investigate the external validity of the segmentation model and its performance in course of time.
Methods
This is a longitudinal study of adult IBD patients guided at three secondary care centres with questionnaires on HRQoL (10-item Short IBDQ, range 10-70) and disease acceptance and control (3-items each, 7-point Likert scale). Two cohorts were created: 1) external validation cohort excluding participants of initial validation study and 2) follow-up cohort of patients with questionnaires after one year. Segments were formed based on mean acceptance and control (cut-off high score>5).
Results
The external validation cohort included 921 patients that were divided in four segments (Figure 1). The acceptance and control scale were unidimensional (85% and 83% of variability explained by the first factor) and internally consistent (Crohnbach’s alpha 0.92 and 0.90). The segments differed significantly in gender, disease duration, IBD medication and clinical disease activity (p<0.05). In multiple regression analysis, high acceptance and high control were significantly associated with a higher HRQoL compared with low acceptance and low control (Beta (95%CI) segment I=11.8 (10.5-13.2), segment II=9.4 (7.8-11.0) and segment III=3.8 (1.6-6.1), p<0.01). The follow-up cohort included 783 patients: 58% remained in the same segment while 42% differed in segment over time (Figure 2). HRQoL differed significantly between patients with positive, negative or no change in segment (p<0.001)(Figure 3). The change in HRQoL over time correlated positively with changes in segment (Spearman rho 0.38, p<0.001) (Figure 4).
Figure 1 Patient segmentation model
Figure 2 Segmentation at baseline and follow-up
Figure 3 Comparison of HRQoL between patients with positive, negative or no change in segment.
* P < 0.001 for baseline HRQoL, follow-up HRQoL and change in HRQoL over time.
Figure 4 Relationship between change in HRQOL and changes in segment over time
Conclusion
This study demonstrated the patient segmentation model based on disease acceptance and perceived control is externally valid and shows consistency over time. The independent association between the different segments and HRQoL was confirmed. In many patients disease acceptance and perceived control remained low after one year. Future research and interventions should aim at improving disease acceptance and perceived control of IBD patients.