Thread by @karmelchoi, Which modifiable factors—diet, physical activity, media use, sleep habits, social interaction, and/or [...]

karmel choi

karmelchoi

Which modifiable factors—diet, physical activity, media use, sleep habits, social interaction, and/or environmental exposures—may be most important for preventing #depression?

We used a 2-stage, genetically informed approach to explore this Q:

https://ajp.psychiatryonline.org/doi/abs/10.1176/appi.ajp.2020.19111158

1/n

Which modifiable factors—diet, physical activity, media use, sleep habits, social interaction, and/or environmental exposures—may be most important for preventing #depression? We used a 2-stage, genetically informed approach to explore this Q: https://ajp.psychiatryonline.org/doi/abs/10.1176/appi.ajp.2020.191111581/n

Using rich data from @uk_biobank, we conducted an "exposure-wide association scan" (ExWAS) across >100 modifiable factors for ~100k individuals, to see which baseline factors were linked to developing depression over time, in general and for those at higher risk.

2/n

What were top associations for incident depression?

: confiding in others, sleep duration, exercises like swimming or cycling, being part of a gym/sports club, etc.

: time spent watching TV or on computer, daytime napping, salt intake, some vit supplements, etc.

3/n

What were top associations for incident depression?: confiding in others, sleep duration, exercises like swimming or cycling, being part of a gym/sports club, etc.: time spent watching TV or on computer, daytime napping, salt intake, some vit supplements, etc.3/n

"But correlation is not causation!" We then used #Mendelianrandomization (MR) analyses to ask, which of these associations are more likely to be causal?

Factors like confiding in others

(risk-reducing), and TV time

(risk-increasing), were still supported.

4/n

"But correlation is not causation!" We then used #Mendelianrandomization (MR) analyses to ask, which of these associations are more likely to be causal? Factors like confiding in others (risk-reducing), and TV time (risk-increasing), were still supported.4/n

Other modifiable factors associated with depression may be a *consequence* of depression (based on suggestive MR findings), or just *co-occur* with depression risks.

e.g., No convincing evidence that taking vitamin B, or eating more cereal, would

depression risk...

5/n

Summary:

We aimed to provide the most comprehensive picture to date of modifiable factors for depression.

Not all associated factors may be causal factors

Systematic approaches can help us refine promising targets for prevention packages.

6/n https://twitter.com/karmelchoi/status/1276135851265851399?s=20