Doctors baffled by Indian village of over 200 sets of twins

[Ivan Seeking]

KODINJI, India (Reuters Life!) – Walk around Kodinji village and you'll think that you have double vision.

The village is home to as many as 230 sets of twins. Nobody knows why there are so many twins in the village of 15,000 people, although one local doctor suspects it might be due to the water.

In fact with about 35-45 twins per live birth [should read: per 1000 live births], this village in North Kerala, India, has four times more twins than normal. Not surprisingly, the village has been dubbed "the twin village."

The latest official estimates by the Kodinji's Twins and Kins Association (TAKA), which conducted door-to-door surveys at the start of the year, found that there were 204 sets of twins...
http://news.yahoo.com/s/nm/20090805/lf_nm_life/us_india_twins

[Evo]

I wonder how much inbreeding there is. How closely related are these people? I saw a documentary about an African village that intermarried within a closed group for so many generations that all of them had the same genetic defect of their feet.

[lisab]

The article doesn't say what percent of the twin births are identical vs. fraternal.

IIRC, fraternal twinning has a genetic component (mother's side only), but identical twinning doesn't. Is this correct?

[CRGreathouse]

Am I missing something? 35 to 45 twin births per 1000 live births gives an expected 525 to 675 (well, 507 to 646 I guess is more proper) sets of twins in a village of 15,000.

Put another way: 204 to 230 sets of twins is something like 13 to 16 twin births per 1000 live births, well below the US rate of ~32.

[Ivan Seeking]

As you said, it appears that we expect a twin birth rate of about 3% [30/1000 births]. In a village of 15,000 we expect about 450 twins, or 225 sets of twins, which is about what is reported.

I'm not sure what to think. Perhaps the problem lies with the current rate of twin births, and not the lifetime running average. But it does sound like they have as many twins as we would expect.

[russ_watters]

Uh, I don't think everyone in the village is having twins! Certainly not the men, the childeren, or the old people! Now that would be something!

I wonder, though, if we look at every village in the world with 15,000 people or more, what the spread of # of twins is. Ie, is it really so surprising to find a village of that size with 4x the normal rate?

[CRGreathouse]

Uh, I don't think everyone in the village is having twins! Certainly not the men, the childeren, or the old people! Now that would be something!

I wasn't assuming that everyone was a mother, but that everyone *had* a mother. (I also implicitly, and wrongly, assumed that life expectancy is the same between twins and non-twins -- but the difference is small, I believe.)

[Ivan Seeking]

I wasn't assuming that everyone was a mother, but that everyone *had* a mother.

Yes, each person alive counts as one live birth, so assuming that twins tend to live as long as non-twins, at any time we would expect approximately 3% of the population to be twins.

[russ_watters]

I wasn't assuming that everyone was a mother, but that everyone *had* a mother. (I also implicitly, and wrongly, assumed that life expectancy is the same between twins and non-twins -- but the difference is small, I believe.) Good point, and I hadn't thought of it. However, I suspect such an analysis would be difficult given the information available. We don't know how many people have entered or left the town, how life expectancy has changed with time or how the twins birth rate has changed with time.

Better in this case to just trust the number they gave is what they say it is. That isn't to say, however, that clarification isn't in order, such as in what timeframe ware we talking about.

[CRGreathouse]

Better in this case to just trust the number they gave is what they say it is.

Are you suggesting that the news reporter understands basic math better than we do? My local (400k circulation) newspaper has 'howlers' almost every day!

[Moonbear]

The incidence of naturally occurring twins is only about 3 in 1000 for Asian populations. The US twinning rate of 30-40 in 1000 is due to the prevalence of assisted reproductive technologies. http://emedicine.medscape.com/article/977234-overview

So, unless a lot of people in that village have been sneaking off for artificial insemination, the twin birth rate they are experiencing is high.

In a small village, I think it is a valid question of how much inter-relatedness there is among the population. Yes, having fraternal twins has a genetic link. An environmental source is also a reasonable possibility, something that is triggering a higher than normal rate of ovulations. An environmental source shouldn't be too hard to identify unless it's a compound not previously known to be an endocrine disruptor. And, there is also that possibility that in a world of statistical averages, some village is bound to be above average by chance alone...but I'm not sure if being that much above average is explainable as just a statistical fluke.