TEMPORAL TRENDS IN QUALITY OF LIFE IN A US COHORT OF PATIENTS WITH CHRONIC PANCREATITIS

Objective. Chronic Pancreatitis (CP) has a detrimental impact on quality of life (QOL). However, with one exception, literature on QOL in CP is limited to cross-sectional data, hindering our ability to understand temporal changes and identify potentially modifiable factors. We evaluated temporal trends in QOL in an ongoing cohort study of CP participants.
Methods. This study included adults with definite CP enrolled in PROCEED, a multicenter, observational cohort study in the USA initiated in 2017. CP was defined by presence of Cambridge 3-4 stage, or parenchymal/ductal calcifications on cross-sectional imaging or histology. PROCEED participants are deeply phenotyped at baseline and annually using structured data collection. Responses to self-administered PROMIS Global Health instrument were used to calculate T-scores for Physical (P-QOL) and Mental (M-QOL) QOL (population mean 50+/-10) and classified into 5 categories (poor-excellent) using established criteria. Multivariable linear mixed models evaluated predictors (demographics, socioeconomic factors, pain and other disease-related factors) for longitudinal QOL. Transition probabilities and longitudinal transition model for QOL categories using data over two sequential visits (years) were used to assess shorter term fluctuations and predictors.
Results. During a median follow-up of 4 years (IQR 2.9-4.9), 516/657 (79%) participants completed at least one follow-up QOL assessment. The mean T-scores for P-QOL and M-QOL at baseline were 43+/-9.3 and 47+/-9.1, respectively. During follow-up, P-QOL or M-QOL trajectories remained stable (p>0.05) (Fig 1) and factors included in multivariable models explained 60% and 47% variability in longitudinal P-QOL and M-QOL scores, respectively. Constant-severe pain, psychological comorbidity and low physical function significantly increased the probability of poorer longitudinal QOL. Other factors with a significant negative impact on QOL were female sex, low education, unemployment and non-married status, smoking, diabetes, new exocrine insufficiency, and disability. Transition probabilities demonstrated notable transitions in P-QOL (26-62%) and M-QOL (35-65%) across the 5 categories at the proximate follow-up year (Fig 2). Participants in very good/excellent categories had a high probability of maintaining QOL, while fluctuations that can be addressed by targeted interventions were more likely in poor-good categories.
Conclusions. Although longitudinal QOL was stable overall, distinct fluctuations within QOL categories were noted for year-to-year evaluation in patients with CP. Our results provide insights into subgroups who may benefit from targeted interventions to improve QOL in CP.