McPherson and colleagues (1986) have suggested that the reason for the discrepant findings is that the studies that showed no association were unable to assess risk after a long potential latent period. As shown in Table B-18, however, four studies have shown that risk is not particularly enhanced more than a decade after exposure to oral contraceptives before a first birth (McPherson et al., 1987; Vessey et al., 1989; Paul et al., 1990) or after 15 years following the first birth in women who used the pill prior to that birth (Schlesselman et al., 1988).

It should also be noted that in two of the studies in Table B-16

TABLE B-15 Relative Risk of Breast Cancer in Women Who Used Oral Contraceptives Before and After Diagnosis of a Benign Breast Lesion

   

Estimate of Relative Risk (Confidence Interval)a

First Author (Date)

Use of Oral Contraceptives

Before Benign Breast Disease

After Benign Breast Disease

CASH b (1986)

Any

0.7(0.5,0.97)

0.8(0.5,1.2)

Stanford (1989)

Any

1.22(0.7,2.1)

0.87(0.6,1.3)

 

<5 years

1.48(0.8,2.7)

0.91(0.5,1.5)

 

>5 years

0.55(0.2,1.7)

0.80(0.4,1.5)

WHO b (1990)

Any

1.30(0.75,2.27)

0.97(0.54,1.72)

Summary relative risk

Any

0.90[0.70,1.16] c

0.86[0.66,1.12] d

a Confidence interval of 95 percent used; [ ] = confidence intervals estimated by approximate methods.

b CASH = Cancer and Steroid Hormone Study; WHO = World Health Organization.

cp value of chi-square test for heterogeneity = .08.

dp value of chi-square test for heterogeneity = .87.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement