Symptom and Age Dependent Casual Effects of Body Size on Bipolar: A Mendelian Randomisation Study

This Mendelian randomisation study reveals that while childhood body size has little causal impact, higher adult body size significantly increases the risk of major depressive disorder, subthreshold mania, and depressive symptoms, suggesting that age-specific body size exerts distinct causal effects on bipolar disorder and related psychiatric features.

The Gist

The Big Picture: The "Body Size vs. Mood" Detective Story

Imagine you are a detective trying to solve a mystery: Does being heavier cause mood disorders like Bipolar Disorder, or does having a mood disorder cause people to gain weight?

For a long time, doctors and scientists have noticed that people with Bipolar Disorder often struggle with their weight. But it's been a confusing puzzle. Is it the weight causing the mood swings? Is it the mood swings causing the weight gain? Or is it just a coincidence caused by other factors (like stress, medication, or shared genetics)?

This study uses a special genetic "detective tool" called Mendelian Randomization (MR) to figure out the true cause-and-effect relationship.

---

The Detective Tool: The Genetic "Randomizer"

Think of your DNA like a lottery ticket you get from your parents. Some tickets have numbers that make you naturally prone to being heavier, while others make you prone to being lighter.

Because you get these tickets at conception (before you are even born), they are like a frozen snapshot of your body size potential. They aren't influenced by your diet, your job, your stress levels, or whether you are currently depressed.

The researchers used these genetic "lottery tickets" to ask: Do people who were genetically destined to be heavier actually end up with more Bipolar Disorder?

If the answer is "yes," it suggests that body size is a cause of the disorder, not just a side effect.

---

The Investigation: Breaking the Problem into Pieces

The researchers realized that "Bipolar Disorder" isn't just one thing. It's like a fruit salad with different ingredients. They decided to look at the ingredients separately:

1. The Subtypes: Bipolar Type I (severe mania) vs. Bipolar Type II (milder mania + depression).
2. The Symptoms: Just the depression part vs. the mania (high energy) part.
3. The Timing: Does being heavy as a child matter? Or does being heavy as an adult matter?

1. The "Child vs. Adult" Twist

The researchers asked: Does the weight you carried when you were 10 years old affect your mental health as an adult? Or is it the weight you carry right now that matters?

• The Finding: Being heavy as a child didn't seem to cause adult mental health issues. In fact, the data suggested that being a bit heavier as a child might even be slightly protective (though this needs more study).
• The Real Culprit: Adult body size was the problem. Being heavier as an adult was strongly linked to:
  • Higher risk of Major Depression.
  • More severe depressive symptoms.
  • More "subthreshold mania" (feeling unusually high or "hyper" but not quite meeting the full criteria for a manic episode).

2. The "Fruit Salad" Confusion

When they looked at Bipolar Disorder as a whole, the results were messy. It looked like body size didn't matter much. Why?

• The Analogy: Imagine you have a scale. On one side, you put Bipolar Type I (which seemed to be slightly less common in heavier people). On the other side, you put Bipolar Type II (which seemed to be slightly more common in heavier people).
• Because these two effects were pulling in opposite directions, they canceled each other out, making the total result look like "zero effect."
• The Lesson: If you treat Bipolar I and Bipolar II as the exact same thing, you miss the hidden differences.

3. The Two-Way Street (Reverse Causation)

The detectives also checked the other direction: Does having Bipolar Disorder make you gain weight?

• The Answer: Yes. The study confirmed that having Bipolar Disorder (and Depression) causes people to gain weight later in life. This makes sense: mood disorders can change your appetite, energy levels, and lifestyle.

---

The Verdict: What Does It All Mean?

Here is the simple summary of the findings:

1. It's a Two-Way Street: Being heavier makes you more likely to get depressed or have mood swings, AND having mood swings makes you more likely to get heavier. It's a vicious cycle.
2. Timing Matters: The weight you carry as an adult is the dangerous part for your mental health. The weight you carried as a child doesn't seem to have the same negative impact.
3. Depression is the Key: The link between body size and mood is strongest with depression. The "mania" part of Bipolar is less clearly linked to weight, and the different types of Bipolar (Type I vs. Type II) might actually react differently to weight.
4. Why This Helps: This tells doctors that treating weight issues in adults might actually help prevent or manage depression and mood swings. It also suggests that we need to stop treating all Bipolar patients as a single group; they might need different treatments based on their specific symptoms and body type.

The "Take-Home" Metaphor

Think of your body and mind like a garden.

• Childhood weight is like the soil quality when the seed was first planted. It didn't seem to ruin the garden later on.
• Adult weight is like the weeds growing now. If you let the weeds (excess weight) take over the garden today, they start choking the flowers (your mood).
• Bipolar Disorder is a storm that can also damage the garden, causing the weeds to grow faster.

The Solution: To keep the garden healthy, you need to pull the weeds (manage adult weight) to stop them from choking the flowers, and you need to protect the garden from the storm (treat the mood disorder) so the weeds don't take over.

Note: This study is a preprint, meaning it is new research that hasn't been fully checked by other scientists yet, but it offers a very clear and exciting new way to look at these problems.

Technical Summary

1. Problem Statement

Previous observational studies suggest a link between higher body size (obesity) and bipolar disorder (BD). However, the causal nature of this relationship remains uncertain due to:

• Confounding: Shared environmental and genetic factors (e.g., lifestyle, medication effects).
• Reverse Causation: Mood disorders may lead to weight gain.
• Heterogeneity: Bipolar disorder is not a monolithic condition; it comprises distinct subtypes (BD-I and BD-II) and symptom profiles (mania vs. depression) that may have different underlying etiologies.
• Age Dependency: It is unclear if body size effects differ based on developmental stages (childhood vs. adulthood).

The study aims to disentangle these factors to determine if body size exerts heterogeneous causal effects on specific bipolar subtypes, symptom profiles, and major depressive disorder (MDD), and whether these effects are age-dependent.

2. Methodology

The study employs a Mendelian Randomization (MR) framework using summary-level data from Genome-Wide Association Studies (GWAS).

• Study Design:

  • Univariable MR (UVMR): Estimates the total causal effect of body size on outcomes.
  • Multivariable MR (MVMR): Simultaneously models childhood and adult body size to isolate age-specific direct effects, accounting for the correlation between early and later life adiposity.
  • Bidirectional MR: Tests reverse causation (effect of MDD/BD on adult BMI).
  • Genetic Correlation: Used Linkage Disequilibrium Score Regression (LDSC) to estimate genome-wide genetic correlations between traits.

• Data Sources:

  • Exposures:
    • Adult BMI: GWAS meta-analysis (n=681,275).
    • Childhood Body Size: Retrospective self-report GWAS (n=453,169) asking about body size at age 10.
    • Adult Body Size: Derived from measured BMI in UK Biobank (n=453,169), categorized to match childhood metrics for comparability.
  • Outcomes:
    • Major Depressive Disorder (MDD), Bipolar Disorder (BD), BD-I, BD-II, Subthreshold Mania, and Depressive Symptom Scores.
    • Data sourced primarily from the Psychiatric Genomics Consortium (PGC) and UK Biobank, with subsets excluding overlapping participants (UKB/23andMe) to minimize bias.
  • Population: European ancestry to avoid population stratification.

• Statistical Analysis:

  • Primary Method: Inverse Variance Weighted (IVW) fixed-effect meta-analysis.
  • Sensitivity Analyses: MR-Egger, Weighted Median, Weighted Mode, MR-PRESSO (for outlier detection), Steiger filtering (to ensure directionality), and Leave-One-Out analyses.
  • Assumptions: Validity of instrumental variables (relevance, independence, exclusion restriction) was rigorously tested.

3. Key Contributions

• Differentiation of Subtypes and Symptoms: Unlike previous studies treating BD as a single entity, this study explicitly separates BD-I, BD-II, and specific symptom dimensions (mania vs. depression).
• Age-Dependent Causal Inference: It is one of the first MR studies to use MVMR to separate the causal effects of childhood body size from adult body size on psychiatric outcomes.
• Bidirectional Analysis: Provides robust evidence for the reverse causal pathway (mood disorders $\to$ body size).
• Resolution of Inconsistencies: Offers a mechanistic explanation for why observational studies show mixed results regarding obesity and bipolar disorder (i.e., opposing effects in BD-I vs. BD-II canceling each other out).

4. Key Results

A. Genetic Correlations

• Adult BMI showed positive genetic correlations with MDD, subthreshold mania, and depressive scores.
• Adult BMI was negatively correlated with Schizophrenia and showed a weak negative correlation with BD-I.
• Childhood body size was positively correlated with adult body size but showed no significant genetic correlation with bipolar outcomes.

B. Univariable MR (Total Effects)

• Adult BMI: Robustly increased the odds of MDD (OR = 1.13), Subthreshold Mania (OR = 1.09), and Depressive Scores.
• Bipolar Disorder: No significant causal effect was found for overall BD, BD-I, or BD-II in the univariable analysis.
• Childhood Body Size: No evidence of a causal effect on any psychiatric outcome.

C. Multivariable MR (Age-Specific Effects)

• Adult Body Size (adjusted for childhood):
  • Significantly increased odds of MDD (OR = 1.37), Subthreshold Mania (OR = 1.63), and Depressive Scores.
  • Significantly decreased the odds of BD-I (OR = 0.76), suggesting a protective effect.
• Childhood Body Size (adjusted for adulthood):
  • Significantly decreased the odds of MDD, subthreshold mania, and depressive scores (suggesting a protective effect of higher childhood body size when independent of adult size).

D. Reverse Causation

• Both MDD and Bipolar Disorder were found to causally increase Adult BMI, supporting a bidirectional relationship between mood disorders and obesity.

E. Sensitivity Checks

• Results were robust across MR-Egger, Weighted Median, and MR-PRESSO methods.
• Heterogeneity was observed, but no consistent evidence of horizontal pleiotropy violated the exclusion restriction assumption for the primary findings.

5. Significance and Discussion

• Mechanistic Insight: The study suggests that the "obesity-bipolar" link observed in observational studies may be driven largely by the depressive component of bipolar disorder (which shares mechanisms with MDD) rather than the manic component. The opposing effects on BD-I (protective) and BD-II (risk-increasing, though weaker) likely cancel out in aggregate analyses.
• Age-Dependency: The findings challenge the notion that childhood adiposity directly causes adult depression. Instead, adult body size appears to be the primary driver of risk for depressive symptoms and subthreshold mania, while childhood body size may have a protective or neutral effect when decoupled from adult weight.
• Clinical Implications:
  • Interventions targeting weight management in adulthood may be crucial for preventing or mitigating depressive symptoms in patients with bipolar disorder.
  • Current diagnostic subtypes (BD-I vs. BD-II) may not fully capture the distinct biological mechanisms regarding metabolic risk; future research should focus on aetiologically defined subgroups.
• Limitations: The study is restricted to European ancestry, limiting generalizability. Power was limited for specific subtype analyses (BD-I/II), and non-linear effects could not be detected.

Conclusion: Body size exerts heterogeneous causal effects on bipolar disorder depending on the life stage and specific symptom profile. Adult body size is a causal risk factor for depressive features and subthreshold mania but may be protective against BD-I, whereas childhood body size does not appear to be a direct causal driver of adult psychiatric outcomes. This highlights the need for age-specific and symptom-specific interventions in mood disorder management.