Continuity of Care for Patients with Chronic Lymphocytic Leukemia: An Analysis of Real-World Data

Introduction

Continuity of care (CoC) is defined as the extent to which healthcare services are received and coordinated, including uninterrupted succession of events consistent with the medical needs of patients. Multiple studies evaluating CoC in solid tumors have shown a strong and statistically significant relationship between greater CoC and reduced health care resource utilization and improved patient outcomes; however, little data exist for hematologic conditions. It was hypothesized that greater CoC would be associated with lower all-cause healthcare resource utilization and improved clinical outcomes among patients diagnosed with chronic lymphocytic leukemia or small lymphocytic lymphoma (CLL) in the United States.

Methods

Optum’s de-identified Clinformatics® Data Mart insurance claims database was used for this study to identify adult patients who were initially diagnosed with CLL between 1/1/2013 and 9/30/2022 with follow up data available through 9/30/2023. Eligible patients were required to have received covalent Bruton tyrosine kinase inhibitor (cBTKi) therapy, have ≥3 unique medical claims in the database from ≥1 hematologist/oncologist, and ≥365 days of continuous enrollment after initiation of cBTKi-based therapy, with an allowable gap of 30 days. Patient-level CoC was evaluated using two published measures: the Herfindahl-Hirschman Index (HHI), a measure of market share dispersion that has been applied to real-world data to quantify consistency, and the CoC Score (CoCS), a measure that further incorporates the number of patient visits to the provider. Exploratory measures of CoC included dispersion of number of providers, usual provider continuity, and two versions of the modified continuity index. All CoC measures are scored from 0.0000 (complete lack of CoC) to 1.0000 (highest possible CoC). Outcomes included all-cause emergency room (ER) visits, inpatient hospitalizations, and overall survival (OS). Multivariable regression models including logistic, Poisson, negative binomial, and Cox proportional hazards were conducted to evaluate the relationship of CoC measures with outcomes, adjusted for baseline covariates of age, sex, race, ethnicity, year of cBTKi treatment, pre-cBTKi systemic therapies received, health insurance, geographic region, time from diagnosis to cBTKi initiation, and cBTKi regimen received.

Results

A total of 4,859 patients were eligible; 60.2% were male, with a median age of 73 (IQR 66-79) years. Most patients were White (81.0%) and most received cBTKi monotherapy (91.4%); 77.5% received a cBTKi in the first-line setting. Median follow-up was 31.8 months. Mean HHI and CoCS values were 0.7040 (SD=0.2643) and 0.6746 (SD=0.2849), respectively. HHI and CoCS were significantly associated with all outcomes evaluated. With every 0.1000 unit increase in CoC with other covariates held constant, there were lower odds of having an ER visit (HHI odds ratio [OR]=0.89; 95% confidence interval [CI] 0.87-0.92, p<0.0001; CoCS OR=0.91, 95%CI 0.88-0.93, p<0.0001) as well as lower number of total ER visits (HHI rate ratio [RR]=0.93; 95%CI 0.92-0.94, p<0.0001; CoCS RR=0.94, 95%CI 0.92-0.95, p<0.0001). Similarly, there were also lower odds of inpatient hospitalization (HHI OR=0.85; 95%CI 0.83-0.87, p<0.0001; CoCS OR=0.86, 95%CI 0.84-0.88, p<0.0001) and lower number of total hospitalizations (HHI RR=0.89; 95%CI 0.87-0.90, p<0.0001; CoCS RR=0.90, 95%CI 0.88-0.91, p<0.0001). Every 0.100 increase in CoC was also associated with lower hazard of death: HHI hazard ratio (HR) 0.98, 95%CI 0.96-1.00, p=0.02; CoCS HR 0.97, 95%CI 0.95-0.99, p=0.001. The interpretation of these findings was consistent in exploratory analyses of the additional measures of CoC.

Conclusions

Greater CoC after initiation of cBTKi-based therapy was associated with reduced ER visits, reduced hospitalization, and improved OS among patients diagnosed with CLL. Interpretation is limited because causality cannot be inferred in the retrospective setting. These findings underscore the importance of maintaining a consistent oncologist/hematologist for the patient over time, particularly as the CLL landscape is rapidly evolving. Future work should explore whether different treatments impact CoC and overall outcomes and should evaluate the timing of provider changes and its impact on outcomes.