Connecting the dots: Network structures of internalizing and functional symptoms in a population-based cohort, 2024, Saini et al

Objective: Comorbidities between internalizing disorders (IDs) and functional disorders (FDs) are well-documented, indicating shared pathways. However, their symptom-level relationships have been largely unexplored. This exploratory study employs a network approach to investigate symptoms of major depressive disorder (MDD), generalized anxiety disorder (GAD), myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), fibromyalgia (FM), and irritable bowel syndrome (IBS) to identify bridge symptoms explaining comorbidity between the two domains.

Methods: We used cross-sectional data on 72,919 adult subjects from the Lifelines Cohort Study, a Dutch general population sample. A total of 38 symptoms representing diagnostic criteria of IDs and FDs were assessed with validated questionnaires. Network models were estimated using eLasso, based on the Ising model, to identify bridge symptoms. The Network Comparison Test (NCT) was used to test whether there were differences in network structure and strength across sex and age.

Results: Symptoms were moderately connected, with a network density of 52.7%. ID and FD symptoms clustered in their respective domains, but were connected through the bridge symptoms, fatigue, difficulty concentrating, trouble sleeping, and unrefreshing sleep. Fatigue and difficulty concentrating had the most connections, associated with 86.6% and 78.9% of the other symptoms, respectively. NCTs indicated no differences in network connectivity between females versus males or younger versus older adults (>50 years).

Conclusions: ID and FD symptoms are moderately interconnected. Bridge symptoms displaying strong connections to multiple disorders may play a central role in the mechanisms underpinning the comorbidity between IDs and FDs.

J Psychosom Res (September 2024) | Link | 10.1016/j.jpsychores.2024.111932 | PDF

 

Common symptoms: common. Groundbreaking stuff.

You could do the same with all diseases and it would come out the same. In a similar fashion, you could randomly throw cards with common symptoms in the air and get similar results.

Or you could do real work with a real purpose. To each their own, I guess.

Also, it's when they mix in things like depression here that they reveal that for the most part, medicine doesn't know a damn thing about it either, can't tell the difference anyway. They truly don't have a single clue about any of this and are randomly flailing and wailing. I doubt that MDD cohorts are any more valid than even the loosest cohorts of 'chronic fatigue'.

 

In this type of symptom-network modelling work, "bridge" symptoms are ones that are commonly endorsed by a wide variety of people irrespective of their specific symptom presentation. Common "bridge symptoms" are sleep problems, fatigue, and concentration problems.

Symptoms might behave in this way because:

1. They are common endpoints of distinctly different causal chains. e.g. a wide variety of problems might lead to sleep difficulties (from depression through to to diabetes and cancer). This idea is called equifinality - entirely different causal pathways lead to an apparently similar outcomes.

2. These symptoms are poorly conceptualised and measured, so people with very different experiences are getting erroneously classed as having the same symptom. e.g., the "poor concentration" that people with severe anxiety experience is entirely different from the "poor concentration" people with COPD experience, but we don't have the conceptual frameworks or language yet to tell these apart.

How do these authors get away with hinting that "bridge" symptoms reflect some sort of common causal mechanism? You just can't make any such claim from this type of modelling.