The cost of doing the right thing
The sepsis bundles, the pressure injury protocols, the VTE prophylaxis workflows, and others represent real progress in clinical care. But every well-intentioned, even important, addition brings three compounding costs along with it.
The additive bias of improvement work
Consider one quarter of improvement work on a typical med-surg unit. Three new evidence-based practices go live: an expanded falls screen, a delirium bundle, and a revised behavioral health assessment. The first adds three minutes per patient per shift. The second adds two. The third adds three. For a nurse covering five patients, that’s 40 minutes per shift—nearly 10% of an eight-hour shift—consumed by additions no one weighed against a fixed time budget.
In 2021, researchers at the University of Virginia published a paper in which, across eight experiments, participants defaulted to additive solutions even when subtractive ones were better. The bias is also self-reinforcing: the more people rely on additive strategies, the more cognitively accessible those strategies become, and subtractive options grow progressively harder to see.
The additive bias is especially strong in healthcare improvement:
- The evidence base only goes one way. Every clinical improvement community produces decades of research on what to add. For instance, Quality Assessment and Performance Improvement (QAPI) programs are designed to solve problems and introduce evidence-based protocols, aligning them squarely with the additive bias. These programs rarely, if ever, weigh the addition against a nurse's finite shift time or the broader portfolio of standard work already in place. Deimplementation science exists as a discipline but has almost no institutional infrastructure in most hospitals.
- The regulatory and accreditation environment rewards adding. Between the Joint Commission, CMS, magnet designation, and payer requirements, nearly every regulator has added fields, screens, and requirements over time. Almost none have systematically retired anything. The compliance ratchet only turns one way.
The accumulation on a modern nursing unit is now substantial. Falls, sepsis, delirium, VTE, pressure injury, nutrition, behavioral health, mobility, and pain screens; CLABSI, CAUTI, VAP, and ABCDEF bundles; the 4Ms framework; hourly rounding with the four Ps; medication reconciliation with teach-back; and discharge teaching—20 to 40 discrete practices expected per patient per shift.
Improvement work rarely focuses on subtraction, and the additive bias is clear.
Standard work and the hidden factory
Additions are only the first cost. Every one of them lands inside existing work systems that leak time on their own.
Behind documented standard work sits a hidden factory of operational failures that nurses absorb to make the standard work happen. Standard work assumes the supplies are stocked, the med is in the Pyxis, the equipment is available, the order is clear, and the information is where it’s supposed to be. The hidden factory is what runs when those assumptions fail.
Anita Tucker and Steven Spear measured this in a 2006 Health Services Research paper, observing nurses minute-by-minute across six U.S. hospitals. Nurses experienced 8.4 operational failures per eight-hour shift, with five categories—medications, orders, supplies, staffing, and equipment—accounting for 76%.
Related research found that only 14% came from employee error; the other 86% were system-generated. A subsequent Kaiser Permanente study found nurses spend 14% of their time working around failures at unmanaged seams between departments: materials, pharmacy, environmental services, biomed, and nursing. Nobody owned the seam.
Across studies, roughly 8% to 14% of every shift is consumed by first-order problem solving, like creating workarounds. On an eight-hour shift, that’s roughly an hour of first-order problem solving; on a 12-hour shift, it’s closer to an hour and a half. The 20-minute wheelchair hunt. The trip to borrow a med. The workaround for the barcode scanner. Each event is small; together they consume nearly a tenth of the shift.
This is a property of the existing work systems, not a byproduct of new additions. There’s already a large pool of recoverable time sitting in these work systems that improvement work rarely touches. And the hidden factory produces two failure modes: time loss when the nurse absorbs the failure through workaround, and reliability loss when the workaround is unavailable and standard work gets skipped or done poorly.
Overprocessing: The stowaway to positive change
When we stand up a new practice, the workflow, EHR build, and documentation get designed under time pressure with many stakeholders each wanting a field added to be safe. The result is over-engineering: the same finding documented in three flowsheets, falls re-screened every shift when the risk score hasn’t changed, signatures collected on forms no one reads, five clicks to record a single assessment, and alerts firing so frequently that the clinically important ones get lost in the noise.
This is overprocessing—waste that consumes time unnecessarily. Duplicative flowsheets, excessive required fields, inconsistent documentation standards, and repeated charting all consume nursing time without necessarily improving care. Streamlining documentation is therefore one of the clearest opportunities to recover time while preserving the clinical intent of the work.
It becomes structural because post-go-live PDSA cycles, if used, often focus on addressing break-fix issues rather than identifying and resolving inefficiencies. The redundant fields, belt-and-suspenders documentation, and over-frequent re-screens become permanent infrastructure. The overprocessing accumulates one implementation at a time and is rarely removed.