12/27/2009

Marriage and Autism: Whatever happened to Journalists Actually Fact-Checking? Updated


Most of this is an older article I wrote on the inaccuracy of the oft-repeated 80% divorce rate.
A new study out of Kennedy Krieger Institute today shows that this statistic is emphatically not correct: “64 percent of children with an autism spectrum disorder (ASD) belong to a family with two married biological or adoptive parents, compared with 65 percent of children who do not have an ASD.” According to the website, researchers came up with these figures from “ data from the 2007 National Survey of Children’s Health[1] , they examined a nationally representative sample of 77,911 children, ages 3 to 17.”
This a myth that is prevalent on the internet and was even repeated by an Autism Speaks presenter at a conference I attended this spring.
One place this myth is presented is at The News on “The Faces of Autism” by Adam Richardson. Granted this is a fairly old piece, dated May 2008, but it’s still relevant, considering the tendency of many writing on autism to engage in quasi-journalism and quoting (and it could be argued the making up of statistics to suit one’s purpose). Richardson writes of a mother with an autistic child: ”Lindblad says more than 80 per cent of marriages with an autistic child end in divorce. She’s living proof.”
Does Richardson ascertain the accuracy of the statistic? Nope. He presents it again at the end of the piece and makes no distinction as to its accuracy:
“FACTS AND MYTHS ABOUT AUTISM
• The divorce rate for autistic parents is over 80 per cent.
• Not all autistic individuals possess special skills and genius abilities. Movies like Rainman created a stigma around the disorder. Less than 10 percent possess “savant” abilities shown in the popular Hollywood movie.
• Not all autistic individuals are withdrawn, avoid eye contact and engage in self-injurious behavior. Many are outgoing and work well within groups of children.”
So, is this 80% a fact or a myth, Mr. Richardson? Do you know or care? Garbage reporting like this seems to be, gasp, epidemic.
Of course, the whole autism ruins marriages and just about everything it touches isn’t new. Kristina Chew wrote more than once on the myth of high divorce rates for parents of autistic children back in 2007.
Lisa  Jo Rudy at Autism.about.com has also touched on it in the past, writing:
“In short, while I don’t honestly believe that 80% of parents with autistic children divorce (because there’s no basis for that figure), I certainly DO believe that 80% (or more) are often under extra stress.” Rudy covers today’s study, noting that Disability Scoop has reported on the study’s findings.
Why do people rely on what research shows for one set of numbers and then pull out another set of numbers with no backing? And the comments are even more illuminating and proof that people rely far too often on the availability heuristic. If it’s easy to recall, then they believe it is true and representative. It’s sort of like the global warming thing: it can’t be real because it’s colder than usual where I live. Folks, there’s a reason scientific evidence trumps anecdote.
*This 80% figure is all over the internet. Susan Leiby throws it out there in an effort to get families assistance by appealing to pity: “80-90% of marriages fail when a child with autism is involved.” Another newspaper uncritically runs with the figure, as well: Denise Ryan in The Age writes with no substantiation: “As well, up to 80% of marriages fail in households where a child has an ASD.”
It keeps going, though. At TACA, Mary Romaniec writes: “The divorce rate in couples who have a child on the spectrum has been suggested to run as high as 80%.”
There seems to be a never-ending parade of folks who like to say statistics say blah blah. I completely get where the Freakonomics authors were coming from as they pointed out the pedophile numbers. Here a minister’s wife passes along the bad information: “Statistics estimate that 80 percent of marriages that have a child with autism fail.” No statistics don’t say that. Another urban legend is born.
This alone ought to make it clear why scientifically-minded folks tend not to rely on Joe Blow saying something. It’s usually bullshit.
And sometimes it’s bullshit that reaches all the way to governmental bodies, like Australia’s Parliament, where Don Randall uses the made-up statistic : “Parents of children with autism face immense challenges. Aside from the care of their child, the resulting stress is enormous, and studies show that up to 80 per cent of
marriages where there is an autistic child fail.” Come on, people! Studies say. Statistics show. Where’s the actual studies, the real statistics? You heard it and it sounds good. It makes the folks who stay married appear all the more heroic and it makes it so much clearer that autism is a burden that ruins lives, right, if 80% of marriages collapse under the weight of autism?
Oh, look what I found: National Autism Association spreading the bull: “The divorce rate in the autism community is estimated at 80%. In an effort to help keep families together, NAA is developing a new program that will provide marriage counseling to parents of children with autism.” I thought they were going to confirm the number before using it?: “NAA is presently conducting a national divorce survey of autism families. Several organizations and news outlets have used the often-quoted autism divorce rate of 80%–NAA hopes to confirm or update that percentage before referencing it in its program materials.” And this was all the way back in June 2007. It’s been 2.5 years; haven’t they figured it out yet? Well, as of this December, they decided to hedge it some and write: “Divorce rates are disproportionately high within the autism community. Government aide is needed for these struggling families.” They are, of course, soliciting donations.
Now, interestingly enough, in the NAA’s letter to Obama, those divorce rates: “Then again, skyrocketing divorce rates in the autism community really need our attention along with the fact that “autism” is just a word some guy came up with 70 years ago to describe a new, rare mental condition that we’re finding is actually more environmental.”
Again, high divorce rates make it so much worse, right? What does all this do? Prove that it’s about the blame-game, victimhood, everything and anything but about the autistic individuals.
Age of Autism’s loyalists buy into the whole thing kit and kaboodle; some of themeven elevating it to 85%.*
What are some realistic estimates of divorce rates?
According to the National Autistic Society, “Siegal (2001) suggests that the divorce rate for couples with a child with autism is the same as that for the rest of the population.”
Kevin Leitch weighed in on the subject this past March, and it’s well worth the read. Leitch discusses the Easter Seals’ survey and its findings that showed that divorce occurred less in parents of children on the spectrum. Leitch also covers the new studyat Left Brain Right Brain.
Sobsey’s (2004) conclusion regarding divorce and parenting disabled children is worth repeating in full:
“In short, evidence for increased marital discord and divorce rates among parents of children with disabilities is weak and inconsistent. Many more parents of children with disabilities report positive effects on their marriages than report negative effects, and many others recognize that having a child with a disability has little to do with the quality or durability of their marriage relationship.
There may be a very small increase in the incidence of divorce among parents of children with disabilities as compared to the general population, or there may be no increase at all. Findings are weak and inconsistent. Even if a small increase in the incidence of divorce exists, it is probably more likely that this increase is attributable to differences in parents’ attitudes and behaviour rather than any effect of children with disabilities on their parents. Whatever the causal factors, many families with children, including many families of children with disabilities, experience marital discord or divorce. Whether or not having a child with a disability is a contributing factor in some cases, marital discord and divorce are difficult for all family members. Researchers should focus future efforts on understanding how children with disabilities and their families experience divorce and what can be done to assist them during what is often a difficult time in their lives.” (p. 80)
I sincerely hope that autism organizations get it right from now on, and that this myth ends. It does a tremendous disservice to families.
(Journal) References:
Siegal, B. (2001). Quality of life: preventing mental separations and legal divorce. What we (don’t) know about the effect of autism on divorce, Advocate, 34(2), pp. 26-29. Available from the NAS Information Centre
Sobsey, D. (2004). Marital stability and marital satisfaction in families of children with disabilities: Chicken or egg?. Developmental Disabilities Bulletin, 32(1), 62-83. Full text available athttp://eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/44/c2/a5.pdf
Web sources have been hyperlinked in the text itself. Siegal’s article was not directly referenced, but the quote taken from the NAS’s page.
(updated to link to Rudy’s coverage of today’s study) May 19, 2010

12 comments:

kathleen said...

"80% or more are under extra stress"? As opposed to what? Does that mean because 3 out of 4 of my kids are on the spectrum-I experience 75% more stress than say a parent with 4 kids but only one on the spectrum? All I can say for sure is that 4 out of 5 dentists prefer Trident...it says so on t.v.

"Grendel" said...

There have been a number of studies about divorce and disability since the 1970s - some using quite small study cohorts (30 families with children who have autism and 30 families with children who do not etc.)

James Rodrigue et al in the Journal of Clinical Child Psychology "Compared mothers of 20 autistic, 20 Down syndrome, and 20 developmentally normal children matched on several pertinent demographic variables. Results revealed that mothers of autistic children reported less parenting competence, less marital satisfaction, more family cohesion, and less family adaptability than mothers in the other two groups."

Which suggests that for many the added work is not breaking families apart necessarily.

Looking at Australian statistics there is certainly a higher number of single parent families where one child has a disability (37% as opposed to around 17% for the general population. Note however that the 17% figure DOES include the subset of families with children with a disability).

A number of studies over the years have shown that for some groups there area actually lower rates of divorce than in the general population - Down Syndrome for example.

I can't access the studies I want to until I get into my office tomorrow but your post has sparked something for me as that 80% figure is one commonly mentioned (without reference) in disability discussions.

I suspect that the 80% figure might actually be an '80% more than something' figure - entirely possible and certainly less than the 'totality' figure most often used.

Thanks for the interesting post (and the two weeks of up-coming literature review!)

KWombles said...

Grendel,

I hope you'll share what you find with us here. Having facts easily at hand to combat the woo comes in handy. :-)

NightStorm The Aspiewolf said...

You know what got me thinking? What about marriages with autistic spouses any data with that? Considering I'm going to be separating with my husband soon seeing the statistics with that sounds interesting.

I kinda feel like writing some tension between Hanai and Maka about Tikaani's issues.

KWombles said...

http://aspergersquare8.blogspot.com/2009/10/daily-squawk-redefining-canines.html Bev's always delightful take on things.

Nightstorm, according to the webosphere, there's supposedly a Dutch study showing an 80% divorce rate for autistic adults. I haven't run that down yet, but am working on it.

"Grendel" said...

KWombles - I will do. It relates to my work so I have a dual interest (I'm also the father of a boy with Autism).

KWombles said...

Posted over in response to Lisa at autism.about.com:

Where are those numbers, Lisa? Isn't it just as bad to throw out generic statements with no support behind them, even if they are lower numbers as it is to throw out the high numbers? This is how myths, inaccuracies and outright lies spread like a wildfire throughout the internet and society.

I spent yesterday digging through academic databases for those numbers to write it up for Countering.
Siegel (2001), according to the NAS, found that it's about the same. Sobsey (2004) also found that it's about the same. The Easter Seals survey from last year showed it's less and that those divorces that did occur were, in general, not because of the child's disability.

Roger Kulp said...

My autisn WAS a major factor in my father getting a divorce...and that was over forty years ago.Only recently,as I have sent him books like Changing tHe Course of Autism has he begun to understand...at 82 years of age.

AutismNewsBeat said...

80 percent of people who quote poorly sourced statistics to make their case can hear 90% of their own peristalsis.

"Grendel" said...

It's been an interesting exercise so far - a 2004 meta-analysis by Risdal and Singer found the following:

"When viewed in terms of the percentage of marriages that end in divorce in the two groups, there is an average increase of 5.97% (range 2.9–6.7%) among families of children with disabilities.

The studies that provide the best evidence based upon large sample sizes and the use of unselected population samples are the studies conducted by Hodapp and Krasner (1995), with a percentage increase in divorce of 5.35%, and by Witt et al. (2003), with an increase of
2.9%. The former study focused on families of children with developmental disabilities and the latter on families of children with a wider range of disabilities, including chronic illness. While these increases are much smaller than previously assumed, they do indicate the
existence of marital difficulties and the need for better forms of family support for some families of children with disabilities."

The meta-analysis did consider some studies specifically with families who have a child with autism but again the sample sizes were very small leading to results that cannot be relied upon.

KWombles said...

Thank you, Grendel, for getting back with what you found.

So we have a fair idea, pooling all this that the divorce may or may not be higher, but if it is, it is not drastically higher.

Supports for families is woefully inadequate. Respite care is nonexistent in many places. Training in how best to parent a child with autism doesn't happen often enough. Support for parents to safely vent their frustrations and gain adaptive coping skills still is charitably mediocre at best.

There's a whole lot of work to be done.

"Grendel" said...

You're welcome - I'm not done yet but all will slow markedly over the new year since I am away from my office with a lot of downloaded PDFs to read!