One In 100 Carry Mutation For Heart Disease

mmmscience brings us news of a new study, published in Nature Genetics by an international team of scientists, that tells a scary story: globally, 1% of the population carry a gene mutation that is almost guaranteed to lead to some form of heart problems. On the Indian subcontinent, the prevalence is 4%. The mutation is a 25-letter deletion of DNA data on the heart protein gene MYBPC3, believed to have arisen in India 30,000 years ago. The researchers say that the mutation wasn't selected out of the population because its effects don't occur until after the childbearing years. The article mentions a prediction that "by 2010 India's population will suffer approximately 60% of the world's heart disease."

Comments

Misleading headline, as usual.

[NereusRen]

Oooh, 1 in 100! Sounds scary! I'm at risk! Wait... lets apply some critical thinking to that number, shall we?

globally, 1% of the population carry a gene mutation that is almost guaranteed to lead to some form of heart problems.

World population is about 6.7B. Total number of people with this mutation in the world:
1% * 6.7B = 67M.

On the Indian subcontinent, the prevalence is 4%.

According to Wikipedia, the subcontinent "accounts for about 40 percent of Asia's population," which is 4B. Total number of people there with this mutation:
4% * 40% * 4B = 64M

So, the percentage of people NOT on the Indian subcontinent that carry this mutation is:
(67M - 64M) / (6.7B - 40%*4B) = 0.06%.

With such a great geographical disparity in incidence, using the global 1% figure to generate the headline of "1 in 100 carry mutation" is incredibly misleading.

The linked article is quite a bit better. It's titled "The heart disease mutation carried by 60 million," and focuses on this as being primarily an Indian problem. Somehow I'm not surprised to see kdawson as the editor on this one.

Re:If the prevalence in India is 4 in 100

[j01123]

I did a quick read of the Nature Genetics letter and, as far as I can tell, it makes no claims as to the worldwide frequency of the allele (actually a micro-deletion). Accurately measuring allele frequencies for the world's population is not something that most studies are adequately designed for, so it's not surprising that they don't provide an estimate. Here's what they have to say about the deletion's frequency outside of India.

The presence of this deletion in many Indian populations with varied geographical and ancestral backgrounds raises the question of how geographically widespread it is outside India. We therefore also analyzed 63 world population samples, comprising 2,085 indigenous individuals from 26 countries including all five continents. The 25-bp deletion was observed in Pakistan, Sri Lanka, Indonesia and Malaysia, (all heterozygotes) but was absent from other samples. Thus, the deletion is a common variant in individuals from South Asia, present in Southeast Asia, but undetectable elsewhere (Fig. 3 and Supplementary Table 5 online).

The supplementary materials give the sample sizes for each of the ethnic groups that were sampled and the number of deletion carriers. Most of the individual samples are small, but in the aggregate they do strongly suggest that the deletion is practically non-existent outside of South Asia and a few neighboring areas.

This does raise the question of how the media got this 1% prevalence estimate, unless I completely missed it in the article. In general, media outlets don't report the contents of peer-reviewed articles, they report the contents of press releases that accompany (or precede) the articles.