Treatment fragmentation and timely treatment in racially and socioeconomically diverse patients with head and neck cancers

Document Type

Conference Proceeding

Publication Date


Publication Title

Cancer Epidemiol Biomarkers Prev


Background: Head and neck cancer treatment is complex, requiring timely, multidisciplinary coordination of care. Integrated care teams, multidisciplinary teams that work together, may improve quality and coordination of care, but can be at odds of patients' ability to access care. Other variables, such as belonging to a racial minority or lower socioeconomic status have also been implicated in access to care. Treatment fragmentation, has been reported to increase cost of care and is associated with worse patient outcomes. However, its effect within a vertically integrated health system, with a salaried medical group, versus private practice, is unknown. This study aims to investigate the risk factors and effects of fragmentation among diverse patients with head and neck cancer treated within a vertically integrated health system.

Methods: This retrospective cohort study investigates diverse patients with head and neck cancer treated from 2012-2019 undergoing ≥2 treatment modalities. Fragmentation was defined as receipt of treatment at >1 treatment facility. Most practice providers practiced out of a single location, here named “Main Campus”. Demographic, disease, and insurance characteristics were collected. Census tract-level socioeconomic status variables include median household income, education composite score, and census block group-level of area deprivation index (ADI) within Michigan. Patients were compared by time to treatment initiation, defined as days from diagnosis to receipt of first treatment modality. In patients undergoing surgery as a first modality of treatment, days from first treatment to second treatment modality was also compared between patients undergoing fragmented and unfragmented care.

Results: Fragmentation occurred in 10.2% of head and neck cancer patients. Unfragmented care was associated with being African American (adjusted odds ratio (aOR)=0.22, 95% confidence interval (CI)= 0.03 to 0.76) and residence in census block with higher ADI (aOR=0.84/decile, 95% CI=0.75 to 0.94). African American patients were more frequently treated at the main campus for their first and second modality of treatments (n=104, 92.9% and n=103, 92.8%) compared to all other races (p<0.001 and p<0.001, respectively). Time to treatment initiation was 34 days (IQR=20, 48) for patients receiving unfragmented care, which was equal to those receiving fragmented care (34 days, IQR=15 to 46; p=0.527). For 158 patients undergoing surgery, time to second treatment modality was also equal between unfragmented (36 days, IQR=29, 44) versus fragmented care (35 days, IQR=30, 41; p=0.624). Neither race nor ADI demonstrated differences in time to treatment initiation or time to second modality treatment.

Conclusion: Within a vertically integrated health system, fragmentation occurred in a small minority of patients. While receipt of unfragmented care was associated with variables traditionally related to decreased access to health care, these populations did not experience delays in initiation of treatment or start of the second modality of treatment.





First Page