Patterns of radiotherapy practice for patients with carcinoma of the uterine cervix: a patterns of care study.

Purpose: To determine the influence of research findings and evolving technology on the practice of radiotherapy in patients with carcinoma of the cervix.

Methods And Materials: Radiation oncology facilities were randomly selected from two strata: those that treated <500 and those that treated >/=500 new cancer patients annually. Patient records were randomly selected from lists of eligible patients treated between 1996 and 1999 at each institution. A total of 442 patient records were reviewed at 59 facilities (55 with eligible patients). National estimates were made using weights that reflected the relative contribution of each institution and of each patient within the sampled institutions. Most estimates were based on the 383 patients who had received their initial therapy at a surveyed facility (excluding the 59 patients who were referred for brachytherapy only after initial treatment at another facility). The estimates of brachytherapy practice were based on the 408 patients who had received their brachytherapy at a surveyed facility. The data were compared with those obtained from a similar 1992-1994 survey.

Results: Overall, 40.5%, 25.4%, and 33.9% of patients had Stage IA-IIA, IIB, or IIIA-IVA disease, respectively. CT was the most common method of nodal evaluation, with surgical evaluation performed in only 12% of patients. Of the 55 facilities with eligible patients, 32 had treated <==8 eligible patients during the 4-year study period. We estimated that 27.5% of patients were treated at facilities that treated <==2 patients with intact cervical cancer annually; 43% were treated at facilities that treated <500 new patients annually. Patients treated at small facilities were significantly more likely to have received a total dose to Point A of <80 Gy, to have had their treatment protracted to >70 days, and to have undergone adjuvant hysterectomy or chemotherapy. In large facilities, radiotherapy was less likely to be protracted to >70 days in the 1996-1999 survey than in the 1992-1994 survey (p < 0.0001); however, in small facilities, treatment was more likely to be protracted than in the earlier survey (p = 0.06), contributing to increasing disparities between the treatments given in large and small facilities. Overall, 92.4% of patients treated with nonpalliative intent were treated with brachytherapy. Of the patients who received brachytherapy, 16.4% had at least part of their brachytherapy delivered at a high dose rate; this was not significantly greater than the proportion treated with high-dose-rate brachytherapy in the 1992-1994 survey (9.1%; p = 0.3). The proportion of patients receiving chemotherapy as part of their initial treatment in 1996-1999 (34.6%) was not significantly different statistically from that in 1992-1994 (25.6%; p = 0.3). However, in 1999, 63% of patients had received chemotherapy compared with 19%, 28%, and 26% in 1996, 1997, and 1998, respectively. The details and confirmation of chemotherapy administration were rarely documented in the radiation oncology clinic notes and hospital records.

Conclusion: The sharp increase in the use of chemotherapy in 1999 suggested rapid application of the results from randomized trials. However, considerable heterogeneity in practice patterns remains, particularly in the use of brachytherapy. The practice at small facilities appears to differ significantly from that at larger facilities in several respects, with a statistically significantly larger proportion of treatments at small facilities failing to meet current guidelines for optimal treatment.