Value Stream Mapping in Healthcare: Current-State to Future-State for a Clinical Pathway

A value stream map is a Lean process-improvement tool, developed for factory floors, that traces every step a unit travels through a process and separates the time that adds value to that unit from the time it just sits waiting. Applied to a hospital, the “unit” moving through the map is not a part on an assembly line — it is a patient, a specimen, or an order, and the process is a clinical pathway: emergency-department triage to inpatient bed placement, a medication order to administration at the bedside, or a specimen collection to a resulted lab value. For a patient-safety officer, quality director, infection preventionist, or risk manager running a process-improvement initiative, that reframing matters: value stream mapping (VSM) gives you a disciplined way to find out exactly where a clinical pathway is losing time, and whether that lost time is genuine waste or a step your program actually needs.

This guide covers what VSM is adapted for a clinical process, how current-state and future-state maps are built, and how to isolate non-value-add wait time specifically — the part of the technique most improvement teams get wrong by mapping the wrong thing or stopping at “we’re just slow” instead of naming the actual wait states.

What Value Stream Mapping Is, Adapted for a Clinical Process

VSM comes out of the Toyota Production System and was formalized for a general audience in Mike Rother and John Shook’s Learning to See (Lean Enterprise Institute, 1998) and popularized more broadly by James Womack and Daniel Jones’s Lean Thinking. The technique’s core move is simple to state and easy to skip in practice: draw every process step as a box, record two numbers for each box — how long the work itself takes (cycle time) and how long the unit waits before the next step starts (wait time) — and total both across the whole pathway. The result separates lead time (the total elapsed time from the first step to the last) from value-added time (the sum of the cycle times that actually change the unit toward its finished state).

A generic process flowchart shows sequence. A value stream map shows sequence plus time at each step plus the information or order flow that triggers each next step — a verbal order, an EHR order entry, a page, a phone call, a physical hand-off. That third element is what makes VSM useful for finding delay: a huge share of clinical wait time is not a task taking too long, it is a signal not being sent, or being sent to the wrong queue, promptly.

Why Patient Flow Needs Its Own Version of VSM

Manufacturing VSM assumes a relatively uniform unit moving through a relatively stable, scheduled line. None of that holds for a clinical pathway, and a team that imports the technique without adjusting for the differences will map something that looks rigorous and isn’t:

• The unit is not uniform. Patients differ by acuity, comorbidity, and case complexity in ways a factory part does not. A current-state map has to either bound itself to a reasonably comparable patient population (e.g., low-acuity ED-to-floor admissions) or explicitly track variance, not just an average.
• Demand is not scheduled. An ED doesn’t have a takt time set by a production schedule; arrivals are largely stochastic. The Lean concept of “leveling” demand still applies (smoothing elective admissions and discharge timing, for example) but has to be adapted rather than copied.
• “Batching” has different stakes. Waiting for a physician to finish rounds before signing a batch of discharge orders is a batching delay in the Lean sense — but the fix isn’t simply “process one at a time,” because rounding itself has clinical value that a factory batch doesn’t.
• Some non-value-add steps are required, not wasteful. Medication reconciliation, informed-consent documentation, and two-person independent double-checks add elapsed time without moving the patient closer to discharge, but they exist for patient safety, not because the process is poorly designed. VSM in healthcare has to distinguish this “required non-value-add” category from pure waste (see below) — collapsing the two leads to a future-state map that cuts a safety check to hit a cycle-time target.

The Institute for Healthcare Improvement’s own framing of Lean adoption in hospitals starts from exactly this reduction: identifying which steps in a process are value-added and which are not, then working the non-value-added ones, as the starting discipline before any specific tool gets applied.

Building the Current-State Map

1. Define the value stream boundary. Pick one trigger event and one end event for a single, well-bounded pathway — “ED bed request placed” to “patient physically in the inpatient bed,” not “everything that happens to an admitted patient.” A map scoped to the whole hospital produces a wall poster nobody can act on; a map scoped to one pathway produces a worklist.
2. Assemble a walking team, not a conference-room team. Include the people who actually touch each step — charge nurse, bed-management/placement coordinator, environmental services, transport, unit clerk — not only their managers. The people closest to a wait state usually already know where it is; the map’s job is to make that knowledge visible and measured.
3. Walk the real process (a Gemba walk), not the policy diagram. Follow an actual case, or a small sample of recent cases pulled from the EHR audit trail, in the order events really happened. A policy flowchart describes the intended process; the current-state map has to describe the one that’s actually running, including the workarounds.
4. Record cycle time and wait time at every step. For each box: how long did the work itself take, and how long did the unit sit idle before the next step started. Timestamp these from the EHR audit trail wherever the system logs them (order placed, order acknowledged, bed assigned, bed cleaned, patient transported) rather than from memory or estimate.
5. Capture the information/order flow between steps. Draw how each step learns that it’s time to start — a discrete EHR order, a page, a phone call, someone walking down the hall to ask. This is where most avoidable delay concentrates, because it’s the part staff don’t think of as “the process” and therefore never fix.
6. Total the map. Sum all cycle times for value-added time; sum the full first-step-to-last-step elapsed time for lead time. The ratio of the two (value-added time ÷ lead time) is the current-state baseline every future-state change gets measured against.

Building the Future-State Map

1. Identify the constraint. On almost every current-state map, the largest opportunity is a wait state, not a task — the longest gap between one step finishing and the next one starting, not the longest single piece of clinical work. Target that gap first.
2. Ask why the gap exists, at the information-flow level. Is the trigger for the next step being sent at all? Is it sent to a shared queue that isn’t actively monitored? Is it sent, but batched until a shift change or a round? Most of the fix lives in this question, not in asking staff to “work faster.”
3. Apply the smallest change that closes the gap, and pilot it. Standard work for a hand-off, a direct notification instead of a routed one, a pull signal (e.g., environmental services notified the moment a discharge order is signed, not after the bed is physically empty) are typical fixes. Avoid redesigning the whole pathway in one pass — a future-state map is a hypothesis, and a large untested redesign is a large untested hypothesis.
4. Draw the future-state map with target cycle and wait times next to the current-state numbers, so the gap being closed is visible on the same page, and attach an implementation plan (a kaizen or rapid-improvement-event plan) with an owner and a date for each change.
5. Re-map after implementation. A value stream map is not a one-time deliverable. Re-walk the pathway and re-time it after the change has been live long enough to be real (not just the first week’s Hawthorne effect), and confirm the gain held before calling the cycle closed.

Identifying Non-Value-Add Wait Time Specifically

The practical skill in VSM is not drawing the boxes — it’s correctly classifying the time inside them. Three categories, not two:

• Value-added time. Work that directly and correctly moves the patient toward the outcome they came in for, done right the first time. A nurse’s clinical assessment, a physician’s exam, medication administration itself.
• Required non-value-added time. Work that doesn’t move the patient closer to the outcome but is legally, regulatorily, or safety-mandated — informed consent documentation, medication reconciliation, an independent double-check on a high-alert medication. This time is a target for how efficiently it’s done, never a target for elimination.
• Pure waste. Time that adds nothing and isn’t required by anything — most wait states fall here. The classic Lean “eight wastes” translate into a clinical pathway roughly as follows: waiting (a patient boarding in the ED for an inpatient bed, a specimen waiting in a rack for pickup, a result sitting unreviewed); transport (moving a patient, specimen, or piece of equipment further than the layout requires); motion (staff walking to a distant supply room instead of a point-of-use stock location); overproduction (duplicate orders placed because the first one wasn’t visible to the ordering clinician); overprocessing (re-documenting the same information in two systems that don’t talk to each other); inventory (excess supply stock, or a patient “boarding” as inventory in a bed they no longer clinically need); defects (rework from a medication error, a mislabeled specimen, or a preventable readmission); and underused skill (a licensed clinician doing work that doesn’t require their license or specific training).

To find the specific wait state at each hand-off, ask one question at every box boundary on the current-state map: what is the patient, specimen, or order actually doing between the moment this step ends and the moment the next one starts? If the honest answer is “nothing, it’s just sitting there,” that gap is pure waste and belongs on the future-state target list. If the answer names a required regulatory or safety step, it belongs in the “required non-value-added” category instead, and the target is efficiency, not removal.

A Worked Example: ED-to-Inpatient-Bed Placement

The figures below are an illustrative composite built to show how the categories apply to a pathway, not measured data from any named hospital, health system, or study.

A current-state map of “ED admission decision” to “patient physically in the inpatient bed” for a low-acuity medical admission might show a lead time of roughly six hours against a value-added time (the actual clinical assessment, order-writing, and hand-off report) of well under an hour. Walking the gap step by step typically surfaces distinct, separately-owned wait states rather than one generic “boarding” problem: a bed-assignment wait (the admission order exists, but no inpatient bed has been assigned in the bed-management system), an environmental-services turnover wait (a bed is assigned but not yet cleaned and released), and a transport wait (the bed is ready, but transport hasn’t been requested or is queued behind other requests). Each of those three has a different owner and a different fix — a pull signal that assigns a bed automatically against unit capacity rather than waiting for a manual review, a housekeeping notification triggered the moment a discharge order is signed instead of after physical bed turnover, and a transport request generated at the same time a bed is marked ready rather than as a separate downstream step. Mapping the pathway as three named waits, each with its own cycle-time and wait-time numbers, is what turns “we’re slow at admissions” into three assignable, testable changes.

Common Pitfalls

• Mapping the policy, not the process. An EHR workflow diagram or a departmental SOP describes the intended pathway. The current-state map has to describe what actually happens, workarounds included, or it will confirm a process that was never really running.
• Scoping too broad. “Map the whole patient journey” produces a map too large to act on. One bounded pathway, with a clear start and end event, produces a worklist.
• Confusing busy with value-added. A step where staff are clearly working hard is not automatically value-added; the test is whether the work moves the patient toward the outcome, not how much effort it visibly takes.
• Treating the map as a one-time poster. Without a re-map after implementation, there’s no way to confirm a future-state change actually held once the initial attention on it fades.
• Cutting required non-value-added steps to hit a target. Compressing or skipping medication reconciliation or an independent double-check to shorten lead time trades a process-improvement win for a safety regression. Required non-value-added time is a target for efficiency, never elimination.

How VSM Fits Alongside Other Quality-Improvement Tools

VSM is a prospective, process-level tool: it maps how a pathway currently runs and designs a target state for it, independent of any single adverse event. That’s a different job from root cause analysis, which is retrospective and event-triggered — RCA asks what caused one specific incident; VSM asks where an entire pathway is losing time, regardless of whether anything has gone wrong yet. The two are complementary: an RCA finding that a delayed hand-off contributed to a deteriorating-patient event is often the trigger for a VSM exercise on that hand-off’s surrounding pathway, and the information-flow analysis inside a value stream map — who signals whom, and how — is close kin to the structured hand-off work covered in SBAR handoff communication. Once a future-state change is implemented, tracking whether the gain held over time is a run-chart or control-chart question, not a VSM question — VSM designs the change; statistical process control monitoring confirms it stuck.

Two implementation guides on this site work through the mechanics of specific pathway redesigns that a VSM exercise commonly produces: barcode medication administration implementation for the medication-pass pathway, and designing a hospital incident-reporting system people actually use for the reporting pathway that often surfaces the wait states worth mapping in the first place.

Frequently Asked Questions

Is value stream mapping the same thing as a process flowchart?

No. A flowchart shows the sequence of steps. A value stream map adds two things a flowchart doesn’t have: a cycle-time and wait-time number at every step, and the information/order flow that triggers each next step. Those two additions are what let a team separate value-added time from wait time and target the wait specifically.

How is a healthcare value stream map different from a manufacturing one?

The unit moving through the process (a patient, specimen, or order) is not uniform the way a manufactured part is, demand isn’t scheduled the way a production line’s is, and some non-value-added steps — consent documentation, medication reconciliation, independent double-checks — are safety-required rather than pure waste. A healthcare VSM has to keep that “required non-value-added” category separate from waste, or the future-state map will target the wrong thing.

Who should be on the mapping team?

The staff who actually touch each step of the pathway — not only their managers. A charge nurse, bed-placement coordinator, environmental services staff, and transport staff typically know exactly where a pathway’s wait states are; the map’s purpose is to make that knowledge visible and measured, not to discover it from scratch in a conference room.

How often should a value stream map be updated?

Re-map after any future-state change has been live long enough to be a real, settled result — not the first week, which usually reflects extra attention rather than a durable fix. Beyond that, treat the map as a living document tied to the pathway’s ongoing improvement cycle, not a one-time deliverable.

Can VSM be used for something other than patient throughput?

Yes. Any pathway with a clear start and end event and multiple hand-offs is a candidate — a lab specimen’s path from collection to a resulted value, a medication order’s path from prescribing to administration, or an incident report’s path from submission to closure all map the same way a bed-placement pathway does.

Sources

• Institute for Healthcare Improvement, Going Lean in Health Care, white paper — on identifying value-added and non-value-added process steps as the starting discipline for Lean adoption in a hospital.
• Mike Rother and John Shook, Learning to See: Value Stream Mapping to Add Value and Eliminate Muda, Lean Enterprise Institute, 1998 — the primary reference for the value stream mapping technique itself, including the current-state/future-state map convention.
• James P. Womack and Daniel T. Jones, Lean Thinking: Banish Waste and Create Wealth in Your Corporation — the general-audience reference for the value-added/non-value-added and “eight wastes” framing this guide adapts to a clinical pathway.