I decided to check out the headlines for the newly released Tamoxifen study. They were predictably embarrassing, and a suitable lesson on why news coverage cannot be considered a trustworthy source on scientific research, especially involving health and medicine:
- “Tamoxifen dramatically lowers incidence of breast cancer in high-risk women, study shows”
- “Tamoxifen linked to lower incidence of breast cancer in high-risk patients”
- “Tamoxifen Drug Can Protect Women With a High Breast Cancer Risk For 20 Years, Scientists Say”
- “Tamoxifen Effective in Lowering Risk of Breast Cancer Recurrence”
- “Tamoxifen could protect women from cancer for 20 years if taken daily”
- “Research shows tamoxifen can cut cancer rates by a third”
That sounds great, but what is being left out?
- The Tamoxifen group had higher rates of cancer than the placebo group at each followup and overall (Incidences of other types of cancers according to follow-up period and treatment allocation).
- The Tamoxifen group had higher mortality rates than the placebo group at each followup and overall (Specific causes of death according to treatment group and follow-up period)
- The Tamoxifen group had more deaths from breast cancer than the placebo group overall (Specific causes of death according to treatment group and follow-up period)
While none of these results is statistically significant it cannot be said that Tamoxifen protected women from death by breast cancer, nor that it allowed them to live longer, nor that it protected them from getting cancer. And this is stated relatively clearly in the paper, but don’t expect the news coverage to pick this up:
Overall there was no significant difference in all-cause mortality between treatment groups, and the excess of death in the tamoxifen group was smaller than in previous reports… Despite there being no significant difference in the number of deaths from breast cancer between treatment groups, there were more deaths from breast cancer in the tamoxifen group… Endometrial cancer was more common in the tamoxifen group than in the placebo group during active treatment, but similar thereafter. However, deaths from endometrial cancer were more frequent in the tamoxifen group (five) than in the placebo group (none), and all but one occurred after 5 years of active treatment. Excess deaths from endometrial cancer have also been reported in the overview of adjuvant tamoxifen trials…”