A blog by Hafsa Benzzi

Causal Threads

Where correlation ends and causation begins. Exploring causal inference across cardiovascular disease, dermatology, genomics, and the boundaries of what observational data can teach us.

H
Dr. Hafsa Benzzi
Dermatologist · PhD Candidate in Clinical Research, Mount Sinai
Latest
Feb 2026 Causal Inference Cardiovascular

The Polypill Paradox: When a Simple Pill Meets Complex Causality

A single pill combining an antihypertensive, a statin, and aspirin. Elegant in concept, messy in real-world evidence. My dissertation uses target trial emulation on 24,788 participants from the All of Us Research Platform to untangle what happens when we move from controlled trials to the chaos of actual patient lives. The results from parametric G-formula, Dynamic Bayesian Networks, and standard regression don't just differ — they tell fundamentally different stories about who benefits and who doesn't.

But what happens when your genes disagree with your treatment plan? That's where PCSK9 variants enter the picture — and where this story takes an unexpected turn.
Continue reading
All Posts
Feb 2026

Target Trial Emulation: The Bridge Between RCTs and Real-World Data

You can't randomize everything. But you can emulate it. Here's how — and where it breaks.

Causal Inference Methodology
Feb 2026

The 30% Problem: Why Biologic Non-Response Demands Causal Thinking

Three in ten psoriasis patients don't respond to biologics. Prediction models tell us who. But only causal models tell us why — and what to do about it.

Dermatology Precision Medicine
Jan 2026

G-Formula vs. Bayesian Networks: A Head-to-Head on the Same Cohort

Same data, same question, different causal architectures. When they converge, you celebrate. When they diverge, that's where the real science begins.

G-Formula Bayesian Networks
Jan 2026

What Dupixent's FDA Journey Taught Me About Evidence Hierarchies

I was a PI on the Phase III dupilumab trials. Eight years later, my PhD is about what RCTs can't tell us. Here's what changed my mind.

Clinical Trials Personal
Jan 2026

DAGs Don't Lie (But They Can Mislead): A Practical Guide

Directed Acyclic Graphs are the lingua franca of causal inference. But drawing the wrong arrow can be worse than drawing none at all.

Tutorial DAGs
Coming Soon — Teasers
01

When Your Genome Overrides Your Prescription

PCSK9 loss-of-function variants modify the effect of statins on cardiovascular outcomes. What does this mean for polypill strategy?

Dropping March 2026
02

Immortal Time Bias: The Silent Killer of Observational Studies

The most dangerous bias is the one you don't see. And it's hiding in most retrospective drug studies — including some landmark ones.

Dropping March 2026
03

From Marrakech to Mount Sinai: Why Cross-Cultural Medicine Needs Causal Methods

Treating psoriasis in Morocco, France, and New York taught me that what works depends on context. Causal models can encode that.

Dropping April 2026
04

The JAK-STAT Paradox in Psoriasis

JAK inhibitors target intracellular signaling. Biologics target extracellular cytokines. What if the right choice depends on your causal phenotype?

Dropping April 2026

Follow the thread

New posts on causal inference, genomics, and the messy space between correlation and causation. No spam. Just science.