Write a note In the chart Compare Glossary Calculators Articles API Get access

Articles

Written by a physician for clinicians: the reasoning behind common decisions, the calculators that support them, and what to ask of any decision-support tool.

Clinical

ASCVD vs PREVENT: which heart risk calculator to use in 2026

The two calculators answer slightly different questions, give different numbers for the same patient, and belong to different sets of treatment thresholds. Mixing them is the error to avoid.

Clinical

Alert fatigue: why most clinical alerts get ignored, and how to design ones that aren't

Clinicians override most alerts, and often they are right to. The fix is not more alerts or louder ones. It is fewer, more specific ones, delivered where the decision is made.

Clinical

Clinical decision support systems: what they are and how to evaluate one

Decision support reliably improves the process of care and has a thinner record on outcomes. Knowing that, and knowing what to ask, is most of what you need to evaluate one.

Clinical

Ruling out PE: Wells, PERC and age-adjusted D-dimer in one workflow

Three tools, one question: can this patient safely skip the CT pulmonary angiogram? Each step answers a narrower question than people assume, and the order matters.

Clinical

MELD 3.0: what changed from MELD-Na and why it matters

The MELD score is still bilirubin, INR, creatinine and sodium at its core. MELD 3.0 adds two variables, two interactions and a creatinine cap, and the reasons for each are worth knowing.

Clinical

Chest pain in the ED: HEART score, the HEART pathway and who can go home

The HEART score is a risk estimate. The HEART pathway is a disposition rule that adds serial troponins. Confusing the two is how a low score turns into an unsafe discharge.

Clinical

Red-flag symptoms: when an outpatient visit becomes an ED visit

Most red flags, taken one at a time, change the probability of serious disease less than we assume. Rules validated as a set, and a clear sense of what is new, do better.

Guide

HIPAA-compliant AI for clinical work: what a BAA chain actually requires

There is no such thing as a HIPAA-certified AI model. There is a chain of contracts, one per hop, and safeguards at each link. Here is how the chain works, with the regulation text.

Guide

OpenEvidence vs UpToDate vs a decision-support API: which does what

Two of these tools help you read and decide. The third hands software structured orders to act on. They overlap less than the comparison suggests.

Guide

UpToDate alternatives in 2026: an honest landscape

Most UpToDate alternatives are other reference libraries, a few are AI answer engines, and one category is not a reference at all. Here is what each does, with prices where they are published.

Clinical

Differential diagnosis generators: what they do, where they help, where they fail

A differential generator answers "what could this be". That is a useful question, and the evidence on how well these tools answer it is more mixed than the marketing.

Guide

AI tools for doctors in 2026: an honest guide by category

The useful question is not which AI tool is best. It is which part of your cognitive load you want to hand off, and what evidence exists that the tool carries it.