A Retrospective Evaluation Of Radiation Therapy Treatment Patterns Among Medicare Enrollees With Head And Neck Cancer

Based on in vitro data and limited clinical data, it is believed that interruptions or early discontinuations in radiotherapy increase the risk of relapse and adversely influence survival. The first step in elucidating the association between deviations from planned radiation treatment and survival of head and neck cancer is to describe patterns of radiation therapy in a large, population-based sample. Our objective was to use Medicare claims data to describe patterns of radiation treatment for persons diagnosed with local or regional head and neck cancer. Using the SEER-Medicare linked database we identified Medicare beneficiaries aged 66 and older diagnosed with local-regional head and neck cancer from 1997 - 2003. We divided the cases into two groups based on treatment: radiation only or surgery + radiation. We calculated the following for each case: 1) time from diagnosis to initiation of radiation; 2) total number of radiation treatments; 3) total duration of radiation; 4) Whether or not a case had a 5-30 day gap in radiation treatment. Among the 4,888 patients who met inclusion criteria, 1,598 (33%) received surgery and radiation, and 3,290 (67%) received radiation only. Among surgical patients, the mean time to initiation of radiation was 71 days (SD = 35.5, median = 66, range = 0 - 180), the mean number of treatments was 31.7 (SD = 8.39, median = 33, range = 2 - 78), and the mean duration of radiation therapy was 54 days (SD = 14.1, median = 53, range = 4 - 166). Among non-surgical patients, the mean time to initiation of radiation was 52 days (SD = 31.7, median = 43, range = 0 - 180), the mean number of treatments was 32.1 (SD = 10.8, median = 33, range = 2 - 115), and the mean duration of radiation therapy was 54 days (SD = 16.0, median = 53, range = 2 - 195). Thirty-seven percent (585) of surgical patients had at least one 5-30 day treatment gap once treatment was initiated, compared to 41% (1355) of non-surgical patients (chi-squared = 9.41, p < 0.01). The time between diagnosis and initiation of radiation therapy was greater for surgical patients compared to non-surgical patients. The range in time between diagnosis and treatment initiation was substantial for both. Substantial proportions of both groups of patients had treatment interruptions but the incidence was slightly higher in the group treated with radiation alone (41% vs. 37%). This large dataset provides a contemporary look at the patterns of care in the management of local-regional head and neck cancer with radiotherapy.