DNA tests are uncovering the true prevalence of incest

[MrPixel]

The "1 in 7000" gets me. I've mentioned before firsthand familiarity with three incest situations - actually four when you count the same father molesting two daughters separately. Three father-daughter, one mother-son.

One father-daughter case resulted in a pregnancy where the child was given-up at birth. Interestingly, that situation was consensual, and it was very disturbing when well after "the incident" I would witness the two greet as adults, and their kiss was not parental. (I was a close friend, obviously, but it did not occur to either their public displays of affection projected TMI.)

The other three were dominance parent-child situations. All four resulted in broken families and broken adults. Two now in their 60s never married, and aside from the "mistake", none have had children.

 

[Kasumi_Lee]

At least my I/T won't be true to life, because no-one is going to get pregnant.

Awkward, my female characters nearly always end up pregnant . . . including in my debut Incest/Taboo story . . .

 

[Bramblethorn]

1 in 7,000 seemed rarer than expected to me, not more common. So I looked up the article in Wikipedia:

https://en.wikipedia.org/wiki/Rare_disease

Most countries classify a disease as rare if its incidence is < 1/2000. So incest is quite rare, even considering that only a fraction of such relationships result in pregnancy.

I think that's usually based on prevalence rather than incidence? They're different measures, and incidence involves a unit of time.

"Incidence" = how often a new case occurs, divided by the population. If everybody in my household catches cold once a year, then the incidence of colds in my household is 100%/year. It might also be measured per-lifetime (if we can expect to live for 80 years, then the incidence might be around 8000% per lifetime) or, particularly for congenital conditions, per live birth.

"Prevalence" = percentage of people who have that condition at any given time. If we get colds once a year, and the average cold lasts three or four days before we fight it off, then on average we spend about 1% of our lives sick with colds, so the prevalence is 1%.

Without being aware of what one's measuring, it's easy to end up making apples-to-oranges comparisons, e.g.:

Thanks, I was going to do the same investigation.

Examples of rare diseases which are more common than 1/7000 include:
ALS (Lou Gherig's/Stephen Hawking's disease)
Mesothelioma
Cystic fibrosis
Scarlet fever
Tay-Sachs disease
Pre-eclampsia

"1 per 7000" is meaningless without knowing "per 7000 what", and changing the "what" can change the number drastically.

(All following numbers referring to the USA, rough numbers only. Some figures are based on numbers that were calculated specifically for white people, so extrapolating to the general population may not be completely accurate, but I'm not trying to be precise here. Also, I'm sleepy and it's quite possible some of my math is wrong, apologies in advance if so.)

Cystic fibrosis is a congenital, lifelong condition. From a population of ~330 million and about 3.6 million births per year, about 1200 people are born with CF per year. So the incidence of CF could be described as "about 1 per 3000 live births", but also as "about 1 per person per 275,000 years".

There are about 30,000 people living with CF in the USA, so the prevalence is about 1 per 10,000 people. (The reason this is different from the "1 per 3000 live births" is that people with CF don't live as long, on average, though that has improved remarkably in the last few decades.)

ALS is something that usually occurs late in life, and people who get it normally only survive a few years. So the incidence is about 1/40,000 to 1/100,000 per year, or about 1/500 to 1/1200 per lifetime, and the prevalence is about 1/16,000.

Pre-eclampsia occurs in about 1/25 of pregnancies in the USA, not particularly rare. But because it "only" lasts about 20 weeks, the prevalence would be about one per 55,000 population (or one per 28,000 women), that being the number who have it at any given time.

Scarlet fever has an incidence of about 1/16,000 per year (~20k cases in the USA annually), or about 1/200 per lifetime, but it usually clears in about a week so the prevalence would be more like 1/800,000.

So whether any of these is "more common than 1/7000" depends on "per 7000 what?" and whether that's comparable to the 1/7000 figure for incest is a whole other question.

 

[MrPixel]

"Lies, damn lies, and statistics."
--Attributed to Mark Twain, but precise origin not known.

 

[Britva415]

whether any of these is "more common than 1/7000" depends on "per 7000 what?" and whether that's comparable to the 1/7000 figure for incest is a whole other question.

We've been talking about prevalence all along.

 

[Bramblethorn]

We've been talking about prevalence all along.

Well, no:

Most countries classify a disease as rare if its incidence is < 1/2000. So incest is quite rare, even considering that only a fraction of such relationships result in pregnancy.

But if we are talking prevalence - leaving aside the question of whether that's the right measure - then I don't think any of the diseases you listed as "rare diseases which are more common than 1/7000" belong in that list.

As discussed above, prevalence is approximately 1/10k for CF, 1/16k for ALS, 1/55k for pre-eclampsia, 1/800k for scarlet fever.

For mesothelioma: USA has about 3000 cases diagnosed a year, but since the average patient survives less than two years after diagnosis, the prevalence would be somewhere below 1/50k.

About the only way Tay-Sachs counts as "more common than 1/7000" is if you're looking at rates at birth, specifically in the Ashkenazi Jewish population. In the general population it's about one in 200k-300k births, and again because the life expectancy is only a few years, the prevalence in the population is much less. In the Ashkenazi population it's about one in 4000 births, but even there the prevalence in the population would be more like 1/60000 because of the life expectancy.

 

[Lifestyle66]

Those incestuous births were probably already a known family secret to the parents. DNA tests are merely exposing the secret facts to others.

Where the surprises come in is for other family members who didn't know they had half-siblings, either raised by a mistress supported by their father, or that their brother/sister is only a half-brother/sister due to "the milkman" fucking their mother.

A neighbor friend I grew up with found when he was 40 years old, he had a half-sister raised by his father's mistress he never knew about. That's why he grew up living in a trailer and their family never seemed to have much money, even though his father was a certified machinist and worked two jobs. The father was paying for two households! And back in those days, a divorced mother with a kid had a tough time finding any other man to marry her and accept her kid. Another neighbor friend our same age was raised by his grandparents to free up his mother to remarry.

Another neighbor recently admitted that at age 60 she learned through the DNA testing site 23 & Me that she had a half-brother her parents never told her about. I've written those facts into at least two of my Loving Wives stories to throw in the face of the "monogamous only" crowd who think older generations were so noble and faithful.

 

[Vitriolhack]

Not passing judgement, just sharing an article:


— The Atlantic (probably behind a paywall)

Emily

I would be very curious to read that article, I wonder if the results indicate recent or past incest activities.

 

[Britva415]

I wonder if the results indicate recent or past incest activities.

https://web.archive.org/web/20240318231550/https://www.msn.com/en-us/news/us/ar-BB1k6mNN

It describes people who are alive today, with first-order incest in their ancestry.

 

[Duleigh]

I'm sorry but 0.01% does not strike me as a pandemic of incestuous romance, even in Scotland

 

[TarnishedPenny]

I'm sorry but 0.01% does not strike me as a pandemic of incestuous romance, even in Scotland

Well, as the article says,

And this number is just a floor: It reflects only the cases that resulted in pregnancy, that did not end in miscarriage or abortion, and that led to the birth of a child who grew into an adult who volunteered for a research study.

So the 1 in 7,000 is misleading. The reality is far higher. How high? That’s the question, isn’t it? It’s a criminal act almost everywhere, a traumatic experience for most of the children involved and every other reason one can imagine it not to discuss or report it.

 

[Duleigh]

Well, as the article says,

So the 1 in 7,000 is misleading. The reality is far higher. How high? That’s the question, isn’t it? It’s a criminal act almost everywhere, a traumatic experience for most of the children involved and every other reason one can imagine it not to discuss or report it.

The problem is that this is a single article by Sarah Zhang. I have seen no actual supporting data. I've read articles all over the web about how DNA tests can show incest but when getting down to the nitty gritty of their findings they say that they got the information from a study by the University of Edinburgh for the UK, but they link back to that single article by Sarah Zhang. And she never links to the study by the University of Edinburgh for the UK, we get that old political pile of horseshit - "Sources say..."

This is supposed to be science, we don't use sources in science, we use data in science. I've seen nothing other than Sarah Zhang's fevered imagination for the source of one in 7,000. which is 0.01%. And by itself that number is too round. PR uses round numbers, Science uses actual numbers. Until actual data can be produced I can only come to the conclusion that she pulled these numbers out of a hat.

 

[TarnishedPenny]

Point taken, Mr D.

 

[SimonDoom]

To be honest, the survey results, even if correct, don't surprise me as either high or low, because I had no preexisting idea of what the numbers would be. It's not something I ever thought about, and it never crossed my mind when I was writing incest stories, because the incest in my stories has nothing to do with the incest that probably occurs in 90+% of the incest cases that show up in the survey (I have no idea what the real percentage is). This data is interesting as evidence of an obviously significant social problem, but it's not interesting to me as an author and has no impact on my feelings as an author of incest stories one way or another.

 

[Bramblethorn]

The problem is that this is a single article by Sarah Zhang. I have seen no actual supporting data. I've read articles all over the web about how DNA tests can show incest but when getting down to the nitty gritty of their findings they say that they got the information from a study by the University of Edinburgh for the UK, but they link back to that single article by Sarah Zhang. And she never links to the study by the University of Edinburgh for the UK, we get that old political pile of horseshit - "Sources say..."

Not just unlinked, it's reported as "unpublished analysis", which presumably means it has not undergone peer review.

DNA testing certainly can give evidence that a person is the child of incest. But translating that into a population rate isn't straightforward, and without the opportunity for others to check the methods used and the assumptions made, I'd be reluctant to depend on those numbers.

I've seen nothing other than Sarah Zhang's fevered imagination for the source of one in 7,000. which is 0.01%. And by itself that number is too round. PR uses round numbers, Science uses actual numbers. Until actual data can be produced I can only come to the conclusion that she pulled these numbers out of a hat.

That part is defensible.

When measuring things, one rarely gets a precise number. Usually we get an estimate with some uncertainty around it. In some fields, the uncertainty is very small; we know the mass of the Earth to within about 0.01%.

In others, the uncertainty can be very large. If my best estimate is a rate of one in 7083 but all I'm sure of is that the true rate is somewhere between one in 6858 and one in 7402 (made-up numbers, I don't know what the actual numbers were)), then reporting that as "one in 7083" risks misleading people as to the accuracy. In that case, rounding to "one in 7000" is fine.

There are fancier ways to report that kind of uncertainty, e.g. "7083 (95% CI 6858 - 7402)", but they're usually not appropriate for a general-audience publication; most people who see that will just latch onto the first number and not know what to do with the rest.

For an estimate of this kind, it's unlikely that they'd be able to get more than one digit's worth of certainty, so "one in 7000" is fine.

 

[Kelliezgirl]

The problem is that this is a single article by Sarah Zhang. I have seen no actual supporting data. I've read articles all over the web about how DNA tests can show incest but when getting down to the nitty gritty of their findings they say that they got the information from a study by the University of Edinburgh for the UK, but they link back to that single article by Sarah Zhang. And she never links to the study by the University of Edinburgh for the UK, we get that old political pile of horseshit - "Sources say..."

This is supposed to be science, we don't use sources in science, we use data in science. I've seen nothing other than Sarah Zhang's fevered imagination for the source of one in 7,000. which is 0.01%. And by itself that number is too round. PR uses round numbers, Science uses actual numbers. Until actual data can be produced I can only come to the conclusion that she pulled these numbers out of a hat.

And of course they pile on the "the reality is worse!" card. Without providing an actual statistical basis for it.
I'd be curious to know to what extent unknown incest plays a part?
In the 50s and 60s informal adoptions were somewhat common in the US. I read a paper about it in a sociology class and they discussed a case where a couple found out they were actually brother and sister when she finally brought him over to meet her parents.