Passing off wrong numbers concerning public health

I first saw the news story this morning in the Times of India mentioning that “postmenopausal women are prone to fractures.” The story highlighted research presented at the European Congress on Osteoporosis & Osteoarthritis. Curiously, the research was also mentioned in a press release by the International Osteoporosis Foundation comparing the prevalence of bone fractures in obese and non-obese women. But this piece is not about the Times of India copying a press release news release nearly word for word: I leave it to you to chew on that revelation. What concerns me is that wrong information is being conveyed to the public.

Osteoporosis is a serious health concern among postmenopausal women, a major segment of the population. So naturally the story interested me. What shocked me was that because of poor math, wrong numbers are being floated to demonstrate the “a high prevalence of obesity in postmenopausal women”.

Consider for a moment the following passage found in both the Times of India story and the original press release from the International Osteoporosis Foundation from which it was copied verbatim:

“A history of fracture after age 45 years was observed in 23 percent of obese and 24 percent of non-obese women. Nearly one in four postmenopausal women with fractures is obese.”

The statement which is being floated is that one out of four (or 25%) postmenopausal women with fractures is obese. That would be abnormally high and a cause for concern. However, you don’t need to know anything about medicine to know the preceding statement that roughly 25% of obese postmenopausal women have absolutely rubbishes it.

Among postmenopausal women, there is only one logical condition which permits the statements that 1) roughly 25% of the obese have fractures and 2) roughly 25% of fractures are in the obese. And that is the number of obese and non-obese people in a population is equal.

Consider what happens if this condition is not met. If there are far fewer obese people in a population, when 25% of those with fractures are obese it means that fractures are much more prevalent in the obese. Really elementary math.

I investigated to see if this was indeed the case. Were there really the same number of obese and non-obese people in Europe? First I found that the definition of obese used by International Osteoporosis Foundation and the World Health Organization were the same (a basal metabolic index of greater than or equal to 30 kg/m2). Next, I looked at a report by the World Health Organization on the percentage of obese women in Europe. Depending on the country in Europe, anywhere from 5% to 20% of women are obese. The number of postmenopausal women in each country varies, but in no European country could the number of obese women approach anywhere near 50% of all those who are postmenopausal.

In fact, if 23% of obese and 24% of non-obese postmenopausal women are afflicted by osteoporosis, the data presented in the study shows that the prevalence is pretty much equal in both groups.

There are many reasons to avoid being obese. Based on the results presented here, however, osteoporosis is NOT one of them.

Of course the wrong headline makes for a better one. Screw the uninformed public.

Advertisements

2 thoughts on “Passing off wrong numbers concerning public health

  1. Its the Times of India. Popularly called TOI-let paper. Of course their data has always been BS. Only now the stench spreads….fortunately.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s