Thursday, July 21, 2005

Bad Headlines on A Breast Cancer Study

The study in question followed women who had had biopsies on breast lumps which were found benign. What the study (covering 9,087 women) suggests is that most benign lumps are not correlated with higher breast cancer risk in the future:

About two-thirds of the women had benign lumps made up of cells that weren't actively growing. The good news is that this most common type of benign tissue didn't increase the risk of developing breast cancer much at all, over an average of ten years after diagnosis.

But, Senay points out, two other types of benign tissue did come with an increased risk. Thirty percent of the women had benign but actively growing cells, and about four percent had atypical, or abnormal-looking cells that were actively growing. Even though they're considered benign, these cells are important to identify because they do elevate the risk of developing breast cancer. They need to be monitored closely and preventive measures could be considered.

And what are the actual risk numbers? According to the study, the average risk is equal to five women in a hundred. The risk in the first of the two groups mentioned above (those with cysts), the one that two-thirds of the study subjects fell into, was six in a hundred.

The risks for two other groups: those where the cells were actively growing though not cancerous and those regarded as atypical were ten and nineteen women in a hundred, respectively. But remember that far fewer women fell into these two groups.

And what does this all mean in ordinary English? Essentially, most benign growths in breast are not associated with any greater likelihood of later breast cancer (the five in a hundred versus six in a hundred difference is most likely not statistically significant), but cancer risk can be higher for certain rarer types of benign growths. The risk factors are simply the relative numbers of women who had been diagnosed with breast cancer during the years that had passed from the benign biopsy in each of the three groups.

I scanned through the Google headlines on this topic, and found everything from:

Most benign breast lumps don't raise risk

which is correct, to many of this type:

Benign breast disease increases cancer risk

The way medical studies are reported in the popular press is often very bad. There really is no good excuse for it.