Du X, Freeman D H, Syblik D A
Department of Internal Medicine and Sealy Center on Aging, University of Texas Medical Branch, Galveston 77555-0460, USA.
Breast Cancer Res Treat. 2000 Jul;62(1):71-9. doi: 10.1023/a:1006414122201.
Three important events in the history of breast cancer treatment occurred between 1983 and 1995: a large clinical trial, first lady Nancy Reagan's choice of mastectomy and the publishing of an NIH consensus statement.
To assess the effects of these events on use of breast conserving surgery (BCS).
Data from the cohort study of the surveillance, epidemiology and end results (SEER) Program from 1983 to 1995 were divided into four periods: Baseline, Trial, Celebrity, and Consensus.
Of the women, 169,466 diagnosed with early stage breast cancer in nine SEER areas.
Monthly percentages of BCS.
A linear regression model generated a separate intercept and slope term for four time periods, adjusting for demographic characteristics of breast cancer patients. For the Baseline, Celebrity and Consensus Periods, slopes indicated an increasing use of BCS which varied between 0.24% and 0.28% per month. Slopes for these three periods were not statistically different (p = 0.120). In contrast, there was no change in use of BCS during the trial period (p = 0.247). We tested the magnitude of discontinuity between periods. At the beginning of the trial, celebrity and consensus periods, there were increases in BCS of 5.54% (p < 0.001), -3.55% (p < 0.001), and 2.37% (p < 0.001), respectively.
The use of BCS was substantially affected by the reports of a clinical trial of BCS and by celebrity action. These effects were abrupt but transient. The NIH consensus statement stimulated a small change in use of BCS and may be an important intervention for maintaining the increasing trend in use of BCS since the 1990s.
1983年至1995年间，乳腺癌治疗史上发生了三件大事：一项大型临床试验、第一夫人南希·里根选择乳房切除术以及美国国立卫生研究院（NIH）共识声明的发布。
评估这些事件对保乳手术（BCS）使用情况的影响。
1983年至1995年监测、流行病学和最终结果（SEER）项目队列研究的数据分为四个时期：基线期、试验期、名人效应期和共识期。
在九个SEER地区被诊断为早期乳腺癌的169,466名女性。
BCS的月度百分比。
线性回归模型为四个时间段分别生成了截距和斜率项，并对乳腺癌患者的人口统计学特征进行了调整。对于基线期、名人效应期和共识期，斜率表明BCS的使用呈上升趋势，每月变化幅度在0.24%至0.28%之间。这三个时期的斜率在统计学上无显著差异（p = 0.120）。相比之下，试验期BCS的使用情况没有变化（p = 0.247）。我们测试了各时期之间的不连续程度。在试验期、名人效应期和共识期开始时，BCS的使用分别增加了5.54%（p < 0.001）、-3.55%（p < 0.001）和2.37%（p < 0.001）。
BCS的使用受到BCS临床试验报告和名人行为的显著影响。这些影响是突然的，但却是短暂的。NIH共识声明促使BCS的使用出现了小幅变化，可能是自20世纪90年代以来维持BCS使用上升趋势的一项重要干预措施。
