Nick Jacobson Profile picture
Oct 27 19 tweets 7 min read
Join me tomorrow at #TIPS2022 for a discussion of Time-Varying Network Dynamics of Major Depressive Disorder

#NetworkScience

Preprint:
psyarxiv.com/pf4kc/

A 🧵for those who can't make it

1/19
Preview: as a field we're making a lot of assumptions that probably aren't right and are kind of a big deal.

2/19
1st assumption:

Specifically:
Changes in MDD are gradual
Symptoms occur "most of the day nearly every day" in major depressive episodes

3/19
Here's what MDD symptoms can actually look like in persons who meet criteria for MDD.

75% of the variation in symptoms occurs within days rather than across days or weeks.

❌Assumption 1 is wrong

MDD symptoms don't change slowly - they change quickly

4/19
2nd Assumption: MDD is a monolithic entity

Only symptom severity matters, the symptoms themselves are interchangeable

5/19
Rather than being some top-down latent process,

When you investigate it MDD symptoms they predict one another over time & do so differently for different people

❌Assumption 2 is wrong

MDD symptoms can function as a network, predicting one another, w/ heterogeneity

6/19
Assumption 3: The MDD network dynamics don't change

The relationships between MDD symptoms are fixed (i.e. anhedonia always has a positive linear relationship to later concentration difficulties for this person)

7/19
So based on prior work, we don't know much about whether MDD symptom dynamics evolve or change across time.

Let's dive in.

8/19
The data from this study were collected as part of an R01, where participants (N = 105 so far) completed measures of their depression sxs 270 times (3x per day for 90 days)

We'll focus on a few participants to dive in detail.

9/19
We ran person-specific (i.e. N = 1) time-varying vector autoregressive models using generalized additive models.

Each change in PHQ symptom is used to predict the changes in each PHQ symptom at the next measurement occasion, & relationships can change over time

10/19
Let's look at the symptom network of participant 1

The symptom changes pretty dramatically over time - concentration goes from having virtually no influence to the most influential in the network

11/19
Participant 2's network is pretty stable for a while until motor symptoms go from essentially carrying no influence to being the most influential symptom in the network

12/19
Participant 3 experiences big changes in how much motor issues influence other MDD symptoms.

13/19
Okay, but what about the whole sample?

86% of participants had their most influential symptom change

ALSO:
*MOST* participants had their most influential symptom change to their least influential symptom at some point in the study.

14/19
So what does this mean for our assumption 3?

It's also wrong ❌

Most people experience dramatic changes in MDD symptom networks across time

15/19
What are the implications of all of this?

MDD is a person-specific and everchanging symptom

This has big implications towards treatments if we try to target these influential symptoms.

16/19
Changes in MDD symptom dynamics change so drastically and so rapidly that targeting central MDD symptoms may be like playing a game of Whac-A-Mole.

The only way to win is to be adapt quickly.

17/19
Probably the only way to meet these rapid symptom network changes is using digital interventions -- traditional interventions are just too slow.

18/19
Thanks to our participants, everyone on the team that has led this work, and all who have inspired this work, including @bringmann_laura, @EikoFried, @aaronjfisher, @BorsboomDenny, & @SachaEpskamp.

19/19

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Nick Jacobson

Nick Jacobson Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @NC_Jacobson

Jan 16, 2020
**Another #tweetorial**

Today, I had another presentation at @SAA2020aus on the importance of asking WHEN examining dynamic in intensive longitudinal data.

This is a summary for those who couldn't make it.

⚠️Warning⚠️
Again, I'm arguing against common practices.

(1)
Intensive longitudinal data allow great opportunity to examine relationships between varying states.

BUT, we have some practices that in this field which are frankly bizarre unscientific heuristics that inform both design & analysis.

Timing is often ENTIRELY overlooked

(2)
In designing active assessment studies when we don't know when an event happens we can use:

Signal-contingent: pinging people & asking to respond usually with either random or windowed random windows

Interval contingent: complete assessments at regular intervals in time

(3)
Read 15 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(