Working Ideas

The Sufficiency Razor

The Sufficiency Razor holds that care should use the lowest level of clinical intensity that can safely and reliably achieve the desired outcome.

J. Michael Eaton · First published September 10, 2026 · Revised November 6, 2026

Question it answers

A healthier population should need less care. Why doesn't that make the care that remains affordable?

Spending moves on two levers: how much illness people develop, and what it costs to deliver care when they need it. Prevention and new therapies such as GLP-1s work on the first. The razor works on the second, which is where most of the gap between U.S. spending and other wealthy countries sits.

How to use it

For every clinical need, ask three questions:

  1. What actually needs to be done?
  2. Who is the lowest sufficient level of clinical expertise?
  3. Where is the lowest sufficient setting in which the need can safely be resolved?

At the front door, the rule is to escalate fast when needed and de-escalate just as fast when the acuity does not hold. Most systems have a well-worn escalation path and no owner for the path back down.

Then change the unit of measurement. Cost per resolved clinical need is the total cost of finishing a patient's need, across every visit, call, and handoff it took, divided by the needs resolved. A need counts as resolved when the patient received what they needed, with no unplanned return for the same problem within 30 days. Report it alongside cost per encounter. Negotiating a $1,000 service down to $800 is progress, and it still misses the $100 pathway that would have resolved the same need.

In practice

At one employed primary care practice, a Capacity Creation Assessment compared the primary diagnosis on nine months of office visits with a physician-sourced list of conditions that can be handled virtually. About one-third of visits qualified, roughly sixty hours per physician per month. Those figures are an identified opportunity. Their value was a number leadership could defend before deciding how much new capacity to build.

Where it breaks

Sufficient is a clinical call.
The razor needs a safety floor set by clinicians. Its failure mode is under-triage, so the front door has to find the high-acuity patient before it routes everyone else down.
A cheaper setting can add care.
In one study of direct-to-consumer telehealth, about 12 percent of visits replaced another visit and 88 percent were new use (Ashwood et al., Health Affairs, 2017). A lower-intensity setting saves money when it replaces the higher one, which takes routing and follow-through.
Fee-for-service pays for intensity.
A system that de-escalates care on fee-for-service revenue gives up margin, which Agility has described as the risk that we can hurt ourselves financially by succeeding clinically. The razor pays off when a payer shares the savings, which is where the Attribute-Steer-Pay test comes in.
The patient has to use it.
The lowest sufficient setting works when it is one the patient will choose, at the hour they need it.

What it borrows

  • Occam's Razor, attributed to William of Ockham: prefer the explanation that needs the fewest assumptions. The Sufficiency Razor applies the same economy to care.
  • The least restrictive setting, a standard in mental health care that asks the same question of placement.
  • Prices over use. Gerard Anderson, Uwe Reinhardt, and colleagues showed that higher U.S. spending comes mainly from higher prices ("It's the Prices, Stupid," Health Affairs, 2003).