A multivariate Swedish national twin-sibling study in women of major depression, anxiety disorder, [FM], and [IBS] 2025 Kendler, Rosmalen et al

[Andy]

Full title: A multivariate Swedish national twin-sibling study in women of major depression, anxiety disorder, fibromyalgia, and irritable bowel syndrome

Abstract

Background
Functional Somatic Disorders (FSD) and Internalizing Psychiatric Disorders (IPD) are frequently comorbid and likely share familial/genetic risk factors.

Methods
We performed a Common Factor Multivariate Analysis of 2 FSDs, Fibromyalgia (FM) and Irritable Bowel Syndrome (IBS), and two IPDs, Major Depression (MD) and Anxiety Disorders (AD), in five kinds of Swedish female–female relative pairs: monozygotic (n = 8,052) dizygotic (n = 7216), full siblings (n = 712,762), half-siblings reared together (n = 23,623), and half-siblings reared apart (n = 53,873). Model fitting was by full information maximum likelihood using OpenMx.

Results
The best-fit model included genetic, shared environmental, and unique environmental factors. The common factor, ~50% heritable with a small shared environmental effect, loaded more strongly on the two IPDs (~0.80) than the 2 FSDs (0.40). Disorder-specific genetic effects were larger for the 2 FSDs (~0.30) than the 2 IPDs (~0.03). Estimated genetic correlations were high for MD and AD (+0.91), moderate between IBS and IPDs (+0.62), and intermediate between FM and MD (+0.54), FM and AD (+0.28), and FM and IBS (+0.38). Shared environmental influences on all disorders were present but small.

Conclusions
In women, FSDs and IPDs shared a moderate proportion of their genetic risk factors, greater for IBS than for FM. However, the genetic sharing between IBS and FM was less than between MD and AD, suggesting that FSDs do not form a highly genetically coherent group of disorders. The shared environment made a modest contribution to the familial aggregation of FSDs and IPDs.

Open access

 

[Utsikt]

Internalizing Psychiatric Disorders (IPD)

First time I’ve heard of this concept. What is it?

 

[rvallee]

I would pretty much assume that doing the exact same thing but based on zodiac signs would yield results that are just as useful as this.

 

[rvallee]

First time I’ve heard of this concept. What is it?

Doesn't really seem to be a thing. Just a vague generic meta category that loosely substitutes for most of the other generic labels and may as well be: "we don't know jack shit but we'll make stuff up to appear otherwise".

Internalizing disorders refer primarily to anxiety and depressive disorders. In the most recent version of the DSM-5, seven distinct anxiety disorders are included: generalized anxiety disorder (GAD), separation anxiety disorder (SAD), social anxiety disorder (social phobia), specific phobia, selective mutism, agoraphobia, and panic disorder. Depressive disorders include disruptive mood dysregulation disorder (DMDD), which is a new diagnosis in the DSM-5, along with major depressive disorder (MDD), persistent depressive disorder, and premenstrual dysphoric disorder.

It would be really interesting if someone could do a study on the numerous labels and made-up terms that form psychosomatic ideology and the dregs of psychiatry. I would be surprised if there are fewer than 50, and they're all either interchangeable at the concept level, or no one can tell the difference anyway.

For example, how does this concept of IPD differ from functional disorders, when either label probably applies to everything they describe. Just mindless nonsense.

 

[Utsikt]

I would pretty much assume that doing the exact same thing but based on zodiac signs would yield results that are just as useful as this.

That was my thought as well. Let’s do zodiac signs and favourite colour next.

 

[Yann04]

That was my thought as well. Let’s do zodiac signs and favourite colour next.

I was thinking what sports team you support and if your bellybutton stick out or in.

 

[Yann04]

In women, FSDs and IPDs shared a moderate proportion of their genetic risk factors, greater for IBS than for FM. However, the genetic sharing between IBS and FM was less than between MD and AD, suggesting that FSDs do not form a highly genetically coherent group of disorders. The shared environment made a modest contribution to the familial aggregation of FSDs and IPDs.

So basically vague made up clinical categories they’ve defined with overlapping symptoms and therefore diagnostic critera have vaguely overlapping genetics?

Who would have thought!

It’s the exact same genuis brain it takes to use diagnostic criteria for ME and Depression that overlap in symptoms and act like it’s a major finding that there is a vague association between ME and Depression.

The fact that this is gold standard psychiatry research is just a massive indigtment of the entire field.

 

[Andy]

First time I’ve heard of this concept. What is it?

Wikipedia says "An internalizing disorder (or internalising disorder) is one type of emotional and behavioral disorder, along with externalizing disorders, and lower incidence disorders.[1] People who have an internalizing disorder will keep their problems to themselves, or internalize the problems."

 

[Hutan]

We collected information on individuals from Swedish population-based registers with national coverage linking each person’s unique personal identification number which, to preserve confidentiality, was replaced with a serial number by Statistics Sweden.

Females only, as there weren't enough males with fibromyalgia

MD, AD, FM, and IBS were defined at the individual level using information on lifetime diagnoses from primary care data and the Swedish population-based registers which include hospital and out-patient specialist-based records (see Supplementary Material, Appendix Table 1).

So lifetime prevalence, up until the time of sampling.

The widely varying sample sizes of our five kinds of relative pairs are seen in Table 1, ranging from 8,052 MZ pairs to 712,762 full sibling pairs. In our pairs from intact families, the lifetime prevalences of MD and AD were ~20%, IBS ~6%, and FM ~1% and moderately higher in our samples from half-sibling families.

That's a huge sample. In the Full Sibs group, which is the biggest group, prevalences in the female Swedish population are
Major depressive disorder 20.8%
Anxiety disorder 19.4%
IBS 6%
Fibromyalgia 1.2%

I'm constantly amazed at the high amount of depression and anxiety disorder diagnoses in general populations around the world. These numbers would be bigger if they were for whole of life. This high background prevalence needs to be kept in mind when interpreting statistics finding that depression and anxiety are common in people with ME/CFS.

 

[Utsikt]

Wikipedia says "An internalizing disorder (or internalising disorder) is one type of emotional and behavioral disorder, along with externalizing disorders, and lower incidence disorders.[1] People who have an internalizing disorder will keep their problems to themselves, or internalize the problems."

Thanks! So we’re still doing Freud..

 

[Hutan]

two Functional Somatic Disorders (FSD) – FM and IBS – and two Internalizing Psychiatric Disorders (IPD) – MD and AD

The authors seem to be completely certain that IBS and fibromyalgia are functional disorders. I thought it was worth noting who they are:

Kenneth S. Kendler 1
Henrik Ohlsson 2
Michael Neale 1
Hanna van Loo 3
Judith Rosmalen 3
Jan Sundquist 2, 4, 5
Kristina Sundquist 2, 4, 5

1: Virginia Institute for Psychiatric and Behavioural Genetics, Virginia Commonwealth University, USA
2. Center for Primary Health Care Research, Lund University, Malmö, Sweden
3. Department of Psychiatry, University Medical Center Groningen, Groningen, Netherlands
4. University Clinic Primary Care Skåne, Region Skåne, Malmö, Sweden
5. Center for Community-Based Healthcare Research and Education (CoHRE), Department of Functional Pathology, School of Medicine, Shimane University, Japan

So, Rosmalen is involved, and people from the US, Sweden, Netherlands. University Clinic Primary Care Skåne looks like a place to avoid. Shimane University in Japan with its Department of Functional Pathology is likely to show up on the functional radar again.

 

[Hutan]

To be clear, they didn't actually do any genetic analysis, or anything particularly complicated at all.

They have these sets of pairs:
Identical twins
Fraternal twins
Fraternal siblings raised together
Half siblings raised together
Half siblings raised apart

Furthermore, using the Swedish national census and population registers, we assessed the cohabitation status for all pairs as the proportion of possible years they lived in the same household until the oldest turned 16.

Just by the way - I find that possibly horrifying - they were able to match up individual health records with census data on cohabitation status. I hope data was deidentified, but surely reidentification would be possible with all of that information?

We fitted a Common Factor model that aims to divide the sources of individual differences in liability to MD, AD, FM, and IBS into additive genetic (A), shared environment (C), and unique environment (E). The model assumes that MZ twins share 100% of their genes; DZ twins and full siblings share, on average, 50% of their genes; while half-siblings share, on average, 25% of their genes. The model also assumes that the shared environment (C), which reflects family and community experiences, is equal between MZ twins, DZ twins, and full siblings, while for half siblings C equaled 1 for pairs reared together and 0 for pairs reared apart. Finally, the unique environment (E) reflects experiences not shared by twins/siblings, random developmental effects, and measurement errors.

So, they are assuming commonality of genetics of 100% for identical twins, 50% for full siblings and 25% for half siblings.
And they give a shared home a rating of 1 for shared environment, and a 0 for a non-shared home. (Their exclusion criteria excluded people who had a mixed history of shared home and not shared home.)

And then they make a model to explain the differences in the prevalence of the 4 disorders in the pairs.

 

[Hutan]

 

[Hutan]

The data in that top panel could be used to support a hypothesis of childhood trauma increasing the incidence of these conditions.
Rates of the conditions are somewhat higher for half siblings (who perhaps faced issues of parental death, divorce or discord) than for full siblings. For example, rates of depression in the full siblings raised together were 20.8% and 28.9% for half siblings raised together. That's probably not surprising for diagnoses of depression and anxiety.

The rate of fibromyalgia in full siblings raised together was 1.2%, but 2.3% in half siblings raised together. The difference for IBS was minimal though (6% for full siblings and 7%/6.4% for half siblings). I didn't see any information given on the statistics - we don't know if these reported prevalences are statistically significant.

The data could also be used to support other hypotheses such as an increased incidence of childhood poverty in families with half siblings, with that poverty directly impacting on things like childhood quality of housing and exposure to pathogens, and indirectly through an increased likelihood of adult poverty acting on those environmental factors.)

 

[Hutan]

Within-disorder cross-relative (the second panel)
(I'm not completely sure what the correlations mean, but I'm assuming they are saying something about how likely the second member of the pair is to have a specified disorder if you know what specified disorder the first member of the pair has.)

The stand out for me there is the enormous implication of a genetic cause of fibromyalgia with a correlation of 0.63 in identical twins. The correlation was only 0.29 in full siblings raised together. The correlation between the depression diagnoses of an identical twin with the depression diagnosis of the second twin was only 0.46 and 0.21 in full siblings raised together.

Just to indicate how much noise there probably is in the data, the correlation in identical twins for IBS was 0.39 and in non-identical twins 0.06. That suggests a significant genetic component and relatively little effect of a shared childhood environment. But the correlation for full siblings raised together was 0.11. Those full siblings could be expected to share a similar amount of genetic variation as the non-identical twins and probably less childhood environmental similarities, and yet the correlation was doubled.

I think there is interesting information here, and would like to hear what people who have looked at a lot of twin studies think about the results.

 

[Hutan]

Cross-disorder within-relative (the third panel)
I think this means, if you know a person has, for example, a diagnosis of depression, what does that tell you about the likelihood of them having, for example, a diagnosis of anxiety. So, the analysis is of the individual.

As would be expected, the correlations are pretty similar across the pair types.

Not surprisingly, there is a big correlation between depression and anxiety diagnoses (around 0.69 in all the pair types). It looks like they are, to a large extent, a package deal. I think that tells us more about the habits of diagnosing doctors and the vagueness of depression and anxiety definitions than anything else.

Correlations between other pairs of disorders all seem to be around the 0.3 mark. I don't think there is much to be made of that. It's a shame that disorders with more obviously physical aetiologies weren't included to give a sense of what relationship we might expect between, for example, depression and a chronic disabling physical condition.

***

It occurs to me that this sort of information is a gold mine for insurance companies, i.e. being able to look at the medical history of a potential applicant for insurance and at the medical history of their siblings and having a very good idea what other diagnoses they are likely to get, and what sort of claims they will make. And then there is actual genetic information.

I imagine academics who provide the data for the ever increasingly detailed modelling of risk will do very well out of support from the insurance industry.

 

[Hutan]

Funding statement
The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. This project was supported in part by a grant from the US National Institutes of Health (R01MH125902) and from the Swedish Research Council (2020-01175 and 2021-06467).

That sounds very vague. Is there any requirement to list out all the sources of funding for an investigation? I wonder to what extent insurers influence decisions about what health investigations are funded with public money.

 

[Hutan]

MadinAmerica article about the first author
Kenneth Kendler: “Implausible” That Psychiatric Diagnoses Even “Approximately True”
I find myself agreeing with him:

In a new article in the top-tier medical journal JAMA Psychiatry, prominent researcher Kenneth Kendler writes that our current psychiatric diagnoses are just “working hypotheses, subject to change.”

According to Kendler, the notion that any psychiatric theories define something “at least approximately true” is “implausible.”

“To argue that our DSM categories accurately correspond to reality would require that among the theories considered was one that was at least approximately true and that the right one was chosen. That is implausible,” Kendler writes. “Given the youth of our science and the complexity of our disorders, it is very unlikely that we now possess definitive theories of their etiology.”
In short, Kendler writes that there is little scientific evidence for psychiatric diagnoses and that he believes the DSM diagnoses do not “correspond to reality” and it is “implausible” that they are “approximately true.”

Kendler is one of the most highly-cited researchers in psychiatry and is famous for his studies on the genetics of schizophrenia. He writes:

“Despite years of research, we cannot explain or directly observe the pathophysiologies of major mental health disorders that we could use to define essential features.”
Kendler makes this argument in favor of what he calls the “instrumentalist” position—that although these diagnoses do not meet any scientific criteria and in fact are unlikely to be “true” in any meaningful sense, they are still sound medical science because he believes in “the reality of major mental illness as an aggregate category.”

Possibly this current study was designed to find that IBS and FM are just variations of depression and anxiety, but I don't think it shows that, even though I suspect the likelihood of a woman who has IBS or FM symptoms getting a depression and/or anxiety disorder label is high.

 

[Utsikt]

MadinAmerica article about the first author
Kenneth Kendler: “Implausible” That Psychiatric Diagnoses Even “Approximately True”

Kendler makes this argument in favor of what he calls the “instrumentalist” position—that although these diagnoses do not meet any scientific criteria and in fact are unlikely to be “true” in any meaningful sense, they are still sound medical science because he believes in “the reality of major mental illness as an aggregate category.”

I struggle to understand this part. It seems to me like this kind of approach would run into massive issues with the differences between correlation and causation? Is that a correct interpretation?

 

[rvallee]

Not surprisingly, there is a big correlation between depression and anxiety diagnoses (around 0.69 in all the pair types). It looks like they are, to a large extent, a package deal. I think that tells us more about the habits of diagnosing doctors and the vagueness of depression and anxiety definitions than anything else.

After years of being subjected to their nonsense, it's painfully clear that all mental health research suffers from the same problem of being studies of what health care systems and clinicians who work in them do, and very little to do with the patients themselves.

This research really says very little about the patients, the illnesses, the conditions. If anything at all. It's not quite random, but is so arbitrary that the day that medical science actually makes the very first breakthrough, they will basically throw away all the old data. Not literally, they will simply not use it at all.

The motivations here are not honest. This is politics, insurance, spendings, and so on. It has almost nothing to do with medicine, health or health care.