Resist the ‘the network is sicker’ conclusion until the data forces you there. Most of the time, it’s something else.

Step 1 — Verify the number.

Did attribution methodology change (members added/removed, new practice acquired)?
Did any upstream data source change (a new hospital feeding claims that wasn’t before)?
Is the denominator (member-months) right, or is the rate inflated by a shrinking denominator?

Step 2 — Segment by acuity.

ED visits by CPT/ICD-resulting admission vs. treat-and-release.
Low-acuity ED visits (the “could have been urgent care” population) up? That’s an access problem, not a clinical-decline problem.
High-acuity up? That’s worth clinical attention.

Step 3 — Segment by member cohort.

New attributed members — often have higher ED use in year 1 before primary care engagement.
Members with chronic conditions vs. generally healthy.
Age bands.
Payer segment — Medicaid vs. commercial vs. Medicare.

Step 4 — External factors.

Flu season severity (CDC ILI surveillance).
COVID resurgence patterns.
Major weather events (heat waves drive cardiovascular ED).
Local factors — closure of a UC clinic, change in hours, reduced PCP capacity.

Step 5 — Access decomposition.

Did PCP appointment wait times get longer? If a member can’t get a PCP appointment for 3 weeks, urgent issues go to the ED. Pull PCP appointment-availability metrics.
Are specific primary care practices driving most of the ED increase? Practice-level heterogeneity is huge.

What “network getting sicker” would actually look like:

High-acuity ED up, low-acuity flat.
Rising ED visits driven by worsening chronic-condition populations.
Paired with rising hospitalization rates, rising per-member-per-month cost, rising mortality.

If it’s just ED without those other signals, you have an access or behavior problem, not a clinical one.

What to tell leadership: “I don’t see the supporting signals of a sicker network. Most of the increase is in low-acuity visits concentrated in two practices that cut evening hours in March. Here’s the proposed intervention and the metric to track it.”

ED visits are up 15%. Is the network getting sicker, or is something else going on?

Draft your answer

Stuck? Peek at a hint

More healthcare questions