Comorbidities in the diseasome are more apparent than real: What Bayesian filtering etc

[Dolphin]

For what it is worth:

Free full text:
http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005487

Abstract
Comorbidity patterns have become a major source of information to explore shared mechanisms of pathogenesis between disorders. In hypothesis-free exploration of comorbid conditions, disease-disease networks are usually identified by pairwise methods. However, interpretation of the results is hindered by several confounders. In particular a very large number of pairwise associations can arise indirectly through other comorbidity associations and they increase exponentially with the increasing breadth of the investigated diseases. To investigate and filter this effect, we computed and compared pairwise approaches with a systems-based method, which constructs a sparse Bayesian direct multimorbidity map (BDMM) by systematically eliminating disease-mediated comorbidity relations. Additionally, focusing on depression-related parts of the BDMM, we evaluated correspondence with results from logistic regression, text-mining and molecular-level measures for comorbidities such as genetic overlap and the interactome-based association score. We used a subset of the UK Biobank Resource, a cross-sectional dataset including 247 diseases and 117,392 participants who filled out a detailed questionnaire about mental health. The sparse comorbidity map confirmed that depressed patients frequently suffer from both psychiatric and somatic comorbid disorders. Notably, anxiety and obesity show strong and direct relationships with depression. The BDMM identified further directly co-morbid somatic disorders, e.g. irritable bowel syndrome, fibromyalgia, or migraine. Using the subnetwork of depression and metabolic disorders for functional analysis, the interactome-based system-level score showed the best agreement with the sparse disease network. This indicates that these epidemiologically strong disease-disease relations have improved correspondence with expected molecular-level mechanisms. The substantially fewer number of comorbidity relations in the BDMM compared to pairwise methods implies that biologically meaningful comorbid relations may be less frequent than earlier pairwise methods suggested. The computed interactive comprehensive multimorbidity views over the diseasome are available on the web at Co=MorNet: bioinformatics.mit.bme.hu/UKBNetworks.

CFS gets a mention.

 

[Dolphin]

https://twitter.com/i/web/status/892402079968448513

 

[Learner1]

Er...they seem to be a bit shy on comorbidities for CFS...

How about these, which seem to be very common on this site and which my expert CFS doctor made a point of testing for?

Hashimotos
Celiac disease
MCAS
Herpes family viruses

I'm sure others here can add to this list.

These types of analyses remind me of the saying "garbage in, garbage out."

 

[SamanthaJ]

Since when has migraine been categorized as a somatic disorder? Or am I misunderstanding the use of the word 'somatic' here? (I know there's another meaning for it, just a bit hazy on what that is)

 

[Esther12]

Since when has migraine been categorized as a somatic disorder? Or am I misunderstanding the use of the word 'somatic' here? (I know there's another meaning for it, just a bit hazy on what that is)

Roughly: 'Somatic' means bodily, 'psychosomatic' means bodily symptoms related to mental processes.

 

[Luther Blissett]

Since when has migraine been categorized as a somatic disorder? Or am I misunderstanding the use of the word 'somatic' here? (I know there's another meaning for it, just a bit hazy on what that is)

soma = greek word for 'body' (i.e. not mind)

somataform = denoting physical symptoms that cannot be attributed to organic disease and appear to be psychogenic.

I get confused around these words too

 

[Jonathan Edwards]

Roughly: 'Somatic' means bodily, 'psychosomatic' means bodily symptoms related to mental processes.

From what I can see they are meaning somatoform or psychosomatic. Arthritis comes up as NOS (not otherwise specified) which is a bit of a joke. I get a whiff of GIGO here.

 

[Esther12]

From what I can see they are meaning somatoform or psychosomatic. Arthritis comes up as NOS (not otherwise specified) which is a bit of a joke. I get a whiff of GIGO here.

We've finally gone full circle with their word games.

 

[CFS_for_19_years]

For what it is worth:

Free full text:
http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005487

Abstract
Comorbidity patterns have become a major source of information to explore shared mechanisms of pathogenesis between disorders. In hypothesis-free exploration of comorbid conditions, disease-disease networks are usually identified by pairwise methods. However, interpretation of the results is hindered by several confounders.

In particular a very large number of pairwise associations can arise indirectly through other comorbidity associations and they increase exponentially with the increasing breadth of the investigated diseases. To investigate and filter this effect, we computed and compared pairwise approaches with a systems-based method, which constructs a sparse Bayesian direct multimorbidity map (BDMM) by systematically eliminating disease-mediated comorbidity relations.

Additionally, focusing on depression-related parts of the BDMM, we evaluated correspondence with results from logistic regression, text-mining and molecular-level measures for comorbidities such as genetic overlap and the interactome-based association score. We used a subset of the UK Biobank Resource, a cross-sectional dataset including 247 diseases and 117,392 participants who filled out a detailed questionnaire about mental health.

The sparse comorbidity map confirmed that depressed patients frequently suffer from both psychiatric and somatic comorbid disorders. Notably, anxiety and obesity show strong and direct relationships with depression. The BDMM identified further directly co-morbid somatic disorders, e.g. irritable bowel syndrome, fibromyalgia, or migraine.

Using the subnetwork of depression and metabolic disorders for functional analysis, the interactome-based system-level score showed the best agreement with the sparse disease network. This indicates that these epidemiologically strong disease-disease relations have improved correspondence with expected molecular-level mechanisms. The substantially fewer number of comorbidity relations in the BDMM compared to pairwise methods implies that biologically meaningful comorbid relations may be less frequent than earlier pairwise methods suggested.

The computed interactive comprehensive multimorbidity views over the diseasome are available on the web at Co=MorNet: bioinformatics.mit.bme.hu/UKBNetworks.

Thanks for posting the abstract.
Posted for the reading-impaired (including myself) who can't read a wall of text.

 

[Karen Kirke]

Here’s a bit more detail on what they say about CFS specifically (I have divided the text into smaller chunks and added bolding):

“Another interesting cluster of comorbid disorders are irritable bowel syndrome (IBS), fibromyalgia (FM), chronic fatigue syndrome (CFS) and migraine; all of them showed strong link with depression based on the text-mining data (S1 Dataset). The BDMM showed that they are strongly co=morbid with depression in the full analysis but when the onset is before depression these strong relationships were absent. (S5 Fig).

The high posteriors in the full analysis, and their sharp decrease in the restricted analysis may indicate that these disorders are heterogeneous themselves: in some subgroups of disorder the symptoms are part of the depression phenotype with high biological overlap but other subgroups maybe independent of depression or adversities that non-specifically predispose to depression.

Similar patterns emerged for insomnia, gastro-oesophageal reflux (gord) / gastric reflux, prolapsed disc/slipped disc, and gastritis/gastric erosions suggesting that in some circumstances they are directly related to depression but in others they are independent of depression (see web tool, Co=MorNet:bioinformatics.mit.bme.hu/UKBNetworks).

When we changed the depression definition to the one defined by low mood and anhedonia, ignoring somatic symptoms [46], only chronic fatigue and fibromyalgia showed co=morbidity with depression, especially with severe depression, suggesting that IBS, migraine, and other above mentioned somatic disorders may have specific relevance for depression dominated by somatic symptoms (for further detail see S3 and S4 Figs and S1 Appendix).”



And another excerpt, where they discuss the implications of the above findings:

“Multimorbidity pattern of IBS, FM, CFS and migraine with depression
“Migraine [34], IBS [68], FM [69], and CFS [70] are highly comorbid with depression based on epidemiologic studies. It is therefore puzzling that they involve different etiological mechanisms. In addition, their symptoms often overlap making it difficult to apply diagnostic categories.

We found that these disorders were not relevant when they occurred before depression but were highly co=morbid in the full analysis. The probable explanation is that in general, these disorders are related to consequences of depression and only specific subtypes of these disorders can be expected to have causal relations, e.g. shared biological background with depression.

For example, a genetic risk score analysis demonstrated that migraine with comorbid depression was more genetically related to depression than to pure migraine, which suggests that migraine might develop as a consequence of different polygenic backgrounds [71]. Similarly, a large general population cohort study confirmed that FM, CFS and IBS increase the odds of depression and anxiety but that most patients who suffer from FM, CFS and IBS have no mood or anxiety disorder [72].”


http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005487#pcbi.1005487.s006

 

[Londinium]

You can play around with the data here: http://bioinformatics.mit.bme.hu/UKBNetworks/

Personally I'm still trying to get my head around the paper and what it's inferring.

 

[SamanthaJ]

Thank you @Esther12 and @Luther Blissett . I think I used to know that. I've been reading too much BPS stuff They've forgotten what 'somatic' means too.

 

[Karen Kirke]

I would have been interested to see how autoimmune conditions would have patterned here.

@Jonathan Edwards, I’d be interested in your interpretation of the quotes from the paper in my post above (some repeated below)?

"Another interesting cluster of comorbid disorders are irritable bowel syndrome (IBS), fibromyalgia (FM), chronic fatigue syndrome (CFS) and migraine; all of them showed strong link with depression based on the text-mining data (S1 Dataset). The BDMM showed that they are strongly co=morbid with depression in the full analysis but when the onset is before depression these strong relationships were absent. (S5 Fig)"

I think this means that there is no real relationship between CFS and depression for people who, at the onset of their CFS, have never been depressed – have I understood that correctly?

"The high posteriors in the full analysis, and their sharp decrease in the restricted analysis may indicate that these disorders are heterogeneous themselves: in some subgroups of disorder the symptoms are part of the depression phenotype with high biological overlap but other subgroups maybe independent of depression or adversities that non-specifically predispose to depression."

It makes a lot of sense to me that a subset of the people who are currently being diagnosed with CFS (or in the case of this study, self-reporting as having CFS) may, in fact, have a condition that overlaps biologically with depression, but that other subgroups have no biological overlap with depression. Treating all subsets the same would lead to ineffective treatments, and potential harm, for the subgroups whose pathophysiologies are not targeted by the treatment.

"We found that these disorders were not relevant when they occurred before depression but were highly co=morbid in the full analysis. The probable explanation is that in general, these disorders are related to consequences of depression and only specific subtypes of these disorders can be expected to have causal relations, e.g. shared biological background with depression."

I find the wording of this bit ambiguous. I am not sure if they are saying that CFS is a consequence of depression (i.e. depression causes CFS) or that depression is a consequence of CFS (i.e. CFS causes depression). While I have come across anecdotes of people who self-report CFS and state that it started with depression, and Naviaux et al (2016 - the dauer paper) were keen to point out that people seemed to end up at the same dysfunctional metabolism regardless of what the trigger for their CFS was, I’d be concerned that we could also be drifting into territory here where chronic fatigue is being confused for CFS. I hope it's awkward wording that could be interpreted to mean the opposite of what they mean, and that what they actually mean is that some people get CFS, and then, afterwards, as a consequence of their CFS, get depression.

 

[A.B.]

Depression is probably way overdiagnosed.

 

[Snow Leopard]

The methodology also comprises a meta analysis of the papers that have been included in the study.
The observed associations also reflect the biases of the design of the studies themselves, eg studies that bothered to measure comorbidities and the types of comorbidities, rather than any underlying association.

I think this means that there is no real relationship between CFS and depression for people who, at the onset of their CFS, have never been depressed – have I understood that correctly?

I think they are overreaching when trying to imply causality given they have not considered the biases associated with diagnosis of depression and CFS - eg patients more likely to get a depression diagnosis before CFS due to doctor ignorance, and patients with a CFS diagnosis first, are very reluctant to get a depression diagnosis afterwards, unless they have severe depression.

 

[Sean]

As with deconditioning, I think the most interesting aspect of ME patients' mental health profile is that it isn't much worse considering the circumstances we have to try to survive in.

 

[Valentijn]

SF36 shows normal mental health for ME patients. So I doubt the studies included are using depression questionnaires which focus on actually feeling depressed, rather than merely being less active.

 

[Jenny TipsforME]

Another factor is that many of us are misdiagnosed as having depression on the diagnostic journey to a ME diagnosis. I don't know how this would be represented.

 

[Woolie]

I think the main problem is that the maps assume that the constructs being measured have some integrity. That they are measuring coherent and non-overlapping things.

If you're depressed because you're ill, this will show up as a link between your illness and depression. Since depression is such a nonspecific measure of distress/physical symptoms - and probably doesn't refer to any single phenomenon - it is likely to be linked with all sorts of other things that cause distress and physical symptoms generally.

Depression is also confounded with social class. So sometimes depression may be indexing bad life situations more generally. And if those bad life situations are also associated with other diagnoses, then that will show up as an association too.

"somatic" is one of those words that looks neutral but is always used to imply psychosomatic.

 

[Woolie]

They say:

The multimorbidity map shows that the well-known psychiatric comorbidities of depression appear to be direct (anxiety, stress, nervous breakdown, postnatal depression etc). Furthermore, the association of several physical disorders such as cardiovascular diseases with depression appears to be indirectly mediated through a direct co-morbidity between obesity and depression.

They used a self-report measure of depression, that looks at things like sadness/loss of joy, negative thoughts about self or future and nonspecific physical complaints like sleep problems, fatigue, poor concentration.

Now see how their statement looks when you substitute the actual thing they did measure whenever you see the word "depression":

The multimorbidity map shows that the well-known psychiatric comorbidities of feeling generally distressed/negative and/or having nonspecific physical symptoms appear to be direct (several other measures that share the general commonality of distress). Furthermore, the association of several physical disorders such as cardiovascular diseases with feeling distressed (etc) appears to be indirectly mediated through a direct co-morbidity between obesity and distress etc.

Doesn't sound all that astounding now, does it?