Sepsis care is the clearest example in nursing of evidence translated into a protocol you can actually execute. A body of trial and observational evidence produced a short, ordered set of actions, the sepsis bundle, and hospitals then measured whether their staff completed it and whether patients lived. That loop, evidence to protocol to measurement to revision, is what evidence-based practice looks like when it works, and it is why sepsis bundles turn up so often in DNP projects and quality improvement work.
The loop just turned again. The Surviving Sepsis Campaign released updated international guidelines in March 2026, superseding the 2021 version, and one change matters more than any other for nurses: qSOFA has been displaced as a screening tool. This guide covers what the 2026 guidelines changed, what the Hour-1 bundle contains, how strong the evidence is for each element, and how to use the bundle as a model for evidence-based practice and quality improvement. For help framing that work into a question you can actually answer, our evidence-based practice and PICOT support is built for it.
Quick Answer:
The 2026 Surviving Sepsis Campaign guidelines now recommend NEWS, NEWS2, MEWS, or SIRS over qSOFA as a single hospital screening tool, a strong recommendation on moderate-certainty evidence. In a comparison across 221,429 prehospital records, NEWS2 achieved a sensitivity of about 73 percent while qSOFA reached only about 23 percent, meaning qSOFA missed roughly three-quarters of sepsis cases. The Hour-1 bundle remains: measure lactate and remeasure if elevated; obtain blood cultures before antibiotics; give broad-spectrum antibiotics; start 30 mL per kilogram of crystalloid for hypotension or a lactate of 4 or more; and add vasopressors to maintain a mean arterial pressure of at least 65 mm Hg. Antibiotic timing is now tiered by probability and shock status rather than applied uniformly. Sepsis remains a clinical diagnosis that no single biomarker can confirm or exclude.
Why the Bundle Is an Evidence-Based Practice Exemplar
Most nursing students meet evidence-based practice as an abstraction: find the evidence, appraise it, apply it. The sepsis bundle makes it concrete. The underlying evidence, that delays in recognition, antibiotics, and resuscitation cost lives, was translated into a fixed sequence of actions with a clock attached. That sequence could then be taught, audited, and improved, which is precisely the translation step that separates evidence-based practice from research.
The distinction matters for anyone planning a project around it. Implementing a sepsis bundle in your unit and measuring compliance is quality improvement. Testing whether a novel screening tool outperforms an existing one is research. Applying an established, guideline-recommended bundle to your patients is evidence-based practice. Those three categories carry different oversight requirements, and confusing them is a common reason DNP proposals stall, which our guide to quality improvement versus research versus EBP works through in detail.
What the 2026 Guidelines Changed
The 2026 Surviving Sepsis Campaign guidelines contain 129 statements, 46 of them new, and were published simultaneously in Intensive Care Medicine and Critical Care Medicine. Several changes affect nursing practice directly.
The screening reversal is the headline. For acutely ill hospitalized adults, the panel now recommends NEWS, NEWS2, MEWS, or SIRS over qSOFA as a single screening tool, a strong recommendation on moderate-certainty evidence. The data behind it is blunt. In a comparative analysis of 221,429 prehospital records, NEWS2 performed best with a sensitivity of 73.1 percent and an area under the curve of 0.77, while qSOFA had the lowest sensitivity at 23.1 percent. A tool that misses roughly three-quarters of cases cannot serve as a screen. qSOFA retains value for flagging clinical deterioration in a patient already suspected of infection, but it is no longer the recommended screening instrument, and any unit protocol still built around it needs revision.
Antibiotic timing is now tiered rather than uniform. For septic shock, antimicrobials should be given immediately, ideally within one hour. For definite or probable sepsis without shock, early administration remains recommended, balanced against diagnostic stewardship. For possible sepsis without shock, the guidance allows a brief window for rapid investigation and antibiotic initiation if concern persists. That tiering reflects a real tension: speed saves lives in shock, but reflexive broad-spectrum antibiotics for every suspected case drive resistance. The 2026 guidelines also recommend against routine empiric anaerobic coverage in patients at low risk, such as most community-acquired pneumonia and urinary tract infections.
Table 1: The Hour-1 Sepsis Bundle and the Strength of Its Evidence
Bundle element | What it requires | Evidence strength | Why it matters |
|---|---|---|---|
Measure lactate | Draw lactate; remeasure if above 2 mmol/L | Suggested: low certainty | Marks tissue hypoperfusion and tracks the response to resuscitation |
Blood cultures before antibiotics | Collect as soon as possible, ideally before antimicrobials | Recommended; low certainty | Preserves yield, so therapy can be narrowed later; supports stewardship |
Broad-spectrum antibiotics | Septic shock: immediately, ideally within 1 hour. Sepsis without shock: tiered by probability | Strong for shock; very low certainty | Speed saves lives in shock; tiering limits unnecessary broad-spectrum use |
Crystalloid resuscitation | 30 mL/kg for hypotension or lactate 4 mmol/L or higher; balanced crystalloid preferred | Suggested: low certainty | Restores perfusion; 2026 guidance adds active fluid removal after resuscitation |
Vasopressors | During or after fluids, to keep MAP at or above 65 mm Hg (60 to 65 if aged 65 or over) | Moderate certainty for the MAP target | Maintains perfusion pressure; may be started peripherally rather than delaying for central access |
Two further changes deserve mention. Balanced crystalloids are now preferred over saline, and the guidelines add an emphasis on active fluid removal once the resuscitation phase has passed, recognizing that fluid overload carries its own harm. For patients aged 65 and over, a lower initial mean arterial pressure target of 60-65 mm Hg is recommended. And in a good practice statement worth quoting to any colleague awaiting a confirmatory test, the panel states plainly that sepsis is a clinical diagnosis that cannot be ruled in or out by a single biomarker.
The Hour-1 Bundle and the Strength of Its Evidence
The bundle itself is unchanged in structure, and knowing the certainty behind each element is what separates a nurse who follows a protocol from one who understands it. Measuring lactate is a suggestion on low-certainty evidence, with remeasurement if the initial value exceeds 2 mmol per liter. Obtaining blood cultures before antimicrobials is a recommendation based on low-certainty evidence, and its rationale is stewardship as much as diagnosis, since cultures drawn after antibiotics have a lower yield and reduce the chance of narrowing therapy later. Broad-spectrum antibiotics follow the tiered timing above. Fluid resuscitation at 30 mL per kilogram for hypotension or a lactate of 4 or more remains, though it too rests on low-certainty evidence. Vasopressors, norepinephrine first, target a mean arterial pressure of at least 65 mm Hg, and the 2026 guidance supports starting them peripherally rather than delaying for central access.
Stating those certainty levels is not hedging; it is the appraisal skill that graduate nursing work demands. A strong recommendation on very-low-certainty evidence, which is how the one-hour antibiotic target for septic shock is graded, represents a deliberate judgment that the lethality of shock justifies acting ahead of definitive proof. Understanding why a panel makes that trade and being able to explain it is exactly what a critical appraisal assignment tests. The same reasoning discipline governs how you rank competing priorities at the bedside, which our COPD care plan walkthrough demonstrates in a different clinical setting.
Table 2: What the 2026 Surviving Sepsis Campaign Guidelines Changed
Area | What changed in 2026 | What it means at the bedside |
|---|---|---|
Screening tool | NEWS, NEWS2, MEWS, or SIRS recommended over qSOFA as a single screening tool (strong, moderate certainty) | Unit protocols built on qSOFA alone need revision; NEWS2 performed best in comparison |
Antibiotic timing | Tiered by shock status and diagnostic probability rather than a uniform target | Immediate in shock; investigation permitted first in possible sepsis without shock |
Anaerobic coverage | Recommends against routine empiric anaerobic coverage in low-risk adults | Target anaerobes only when indicated, such as bowel perforation or necrotizing infection |
Fluids | Balanced crystalloids preferred over saline; new emphasis on active fluid removal after resuscitation | Resuscitate, then de-resuscitate; fluid overload is treated as its own harm |
Blood pressure target | Lower initial MAP target of 60 to 65 mm Hg suggested for adults aged 65 and over | Age-adjusted target rather than one threshold for all adults |
Quality improvement | New recommendation for systematic, ongoing improvement of care processes; code sepsis or sepsis huddle protocols suggested | Implementation work is now explicitly part of the guideline, not an optional extra |
The Nurse's Role in Bundle Delivery
Almost every element of the bundle passes through nursing hands, which is why bundle compliance is so sensitive to the nursing process. The triage or ward nurse is usually the first to detect deterioration, which is what makes the screening tool change consequential rather than academic. Under the 2026 guidance, a unit screening with qSOFA alone is the least sensitive available instrument.
From recognition onward, the sequence is nurse-executed: drawing lactate and cultures, ensuring cultures precede antibiotic administration, hanging the antibiotic without waiting for a bed or a transfer, delivering the fluid volume as a rapid infusion rather than at a maintenance rate, and escalating to vasopressors when pressure does not respond. Reassessment follows: mean arterial pressure, mentation, capillary refill, urine output, and repeat lactate. Documentation carries real weight here, because time zero, the moment sepsis was first documented, anchors every compliance measure that follows, and imprecise charting can make compliant care look non-compliant; our clinical documentation support addresses that recording discipline directly.
Care planning for the septic patient follows the same prioritization logic as any acute deterioration, with tissue perfusion and fluid volume leading, and the education and prevention diagnoses arriving later. If you are building that plan formally, our nursing care plan support covers the structure.
The ranking discipline is identical to the one our DKA care plan walkthrough applies to a metabolic emergency, where the intervention everyone reaches for first is not the one that should lead.
Sepsis Bundles as Quality Improvement Projects
The 2026 guidelines add something new and directly useful to anyone planning a project: an explicit quality improvement recommendation, describing a systematic and ongoing effort to improve evidence-based care processes, including timely antibiotic administration, and suggesting a code sepsis or sepsis huddle protocol to accelerate diagnosis and treatment. That is a guideline body telling you, in effect, that implementation work is part of the intervention.
The evidence that bundle compliance tracks with survival is substantial. New York State's mandated sepsis protocols produced the clearest dataset: across more than 91,000 patients at 183 hospitals, three-hour bundle compliance rose from 53.4 percent to 64.7 percent while risk-adjusted mortality fell from 28.8 percent to 24.4 percent. Analyses of the federal SEP-1 measure have similarly associated compliance with lower 30-day mortality, and time-to-completion analyses have found each additional hour associated with higher mortality.
Those findings should be presented carefully, and doing so is itself a mark of graduate-level work. They are observational associations, not randomized proof. Hospitals that achieve high compliance differ from those that do not in staffing, resources, and culture, and a Hawthorne effect is plausible under mandated reporting. The Infectious Diseases Society of America declined to endorse the earlier bundle due to concerns about antibiotic overuse in patients who do not have sepsis. The benefit is most robust in septic shock and weakest in possible sepsis without shock, which is precisely why the 2026 guidelines tiered the timing. A DNP project that states this plainly is stronger than one that treats bundle compliance as settled causation.
Appraising the evidence is where most sepsis projects lose marks |
|---|
Committees rarely reject a sepsis project for choosing the wrong topic. They reject it for overstating what the compliance data proves, for classifying the work incorrectly, or for a question that cannot be answered with the data available. Send us your proposal or appraisal, and a reviewer will pressure-test the question, the evidence claims, and the framework before your committee does. Get your evidence-based project reviewed, with an itemized quote within 2 to 4 business hours, no obligation. |
For a project, the structure almost writes itself. Compliance is the process measure, time to antibiotics and time to fluid completion are balancing measures, and risk-adjusted mortality or length of stay is the outcome. Framing that as an answerable question is the first real hurdle, and our guide to writing a PICOT question sets out how to build one that survives committee review.
Choosing the implementation model that structures the rollout is the next decision, and our comparison of the Iowa, Johns Hopkins, and JBI frameworks covers how to pick one and defend it in front of a committee.
Projects of this shape, a guideline-recommended bundle implemented and measured in a single unit, are among the most common we support through our DNP project service.
Common Mistakes
A few errors recur in both practice and student work. The first is continuing to screen with qSOFA alone, now explicitly not recommended. The second is treating the bundle as five independent tasks rather than an ordered sequence, most damagingly by sending antibiotics before cultures are drawn, which forfeits the chance to narrow therapy. The third is stating bundle-mortality findings as causal when they are associational. The fourth is applying the one-hour antibiotic target uniformly, when the 2026 guidance deliberately tiers it by shock status and diagnostic probability. The fifth, in project work, is misclassifying a bundle implementation as research when it is quality improvement, or the reverse.
The corrective across all of these is the same discipline: read what the guideline actually recommends, note the certainty attached to each recommendation, and describe the evidence at the strength it actually holds. That is what evidence-based practice asks, and the sepsis bundle is a good place to learn it, because the stakes make the reasoning visible.
Frequently Asked Questions
What is in the Hour-1 sepsis bundle?
Five elements: measure blood lactate and remeasure if it exceeds 2 mmol per liter; obtain blood cultures before administering antimicrobials; administer broad-spectrum antibiotics; begin rapid administration of 30 mL per kilogram of crystalloid for hypotension or a lactate of 4 mmol per liter or higher; and apply vasopressors during or after fluid resuscitation to maintain a mean arterial pressure of at least 65 mm Hg. The bundle is initiated on recognition, and not every element will always be completed within one hour.
Why did the 2026 guidelines stop recommending qSOFA for screening?
Because of sensitivity. In a comparative study of 221,429 prehospital records, qSOFA had the lowest sensitivity of the tools examined at 23.1 percent, meaning it missed roughly three-quarters of sepsis cases, while NEWS2 achieved 73.1 percent with an area under the curve of 0.77. The 2026 guidelines therefore recommend NEWS, NEWS2, MEWS, or SIRS over qSOFA as a single screening tool, a strong recommendation on moderate-certainty evidence. qSOFA remains useful for identifying deterioration in a patient already suspected of having an infection.
How quickly should antibiotics be given in sepsis?
Timing is now tiered. For septic shock, antimicrobials should be given immediately, ideally within one hour. For definite or probable sepsis without shock, early administration is recommended but balanced against diagnostic stewardship. For possible sepsis without shock, a short period of rapid investigation is acceptable, with antibiotics given if concern persists. The 2026 guidelines also advise against routine empiric anaerobic coverage in patients at low risk of anaerobic infection.
What is the difference between sepsis and septic shock?
Under the Sepsis-3 definitions, sepsis is life-threatening organ dysfunction caused by a dysregulated host response to infection, operationalized as an acute rise of two or more points in the SOFA score. Septic shock is a subset of sepsis in which circulatory and metabolic abnormalities are severe enough to require vasopressors to maintain a mean arterial pressure of at least 65 mm Hg, together with a lactate level above 2 mmol per liter, despite adequate fluid resuscitation. Septic shock carries substantially higher mortality.
Is a sepsis bundle project quality improvement or research?
Implementing an established, guideline-recommended bundle and measuring compliance and outcomes in your own setting is quality improvement, and it is what most DNP sepsis projects are. Testing a new screening tool or comparing interventions to generate generalizable knowledge is research and carries different oversight requirements. The distinction determines your review pathway, so settle it before writing the proposal rather than after.
You May Also Find Useful
Evidence, Then Protocol, Then Measurement
The sepsis bundle earns its place in nursing education because it shows the full arc of evidence-based practice rather than one slice of it. Evidence identified the time-sensitive interventions. A guideline body turned them into an ordered protocol with certainty ratings attached. Hospitals measured compliance and outcomes, and those measurements fed back into a revised guideline, which is exactly what happened in 2026 when qSOFA was displaced by more sensitive screening tools. If you are building a project on this, check the current guideline rather than an older summary, state each recommendation at the certainty it carries, and be candid that the compliance-mortality relationship is an association rather than proof. That honesty is not a weakness in the argument; it is the argument.
Building a sepsis bundle project or an evidence-based practice assignment, and want the question and appraisal checked before you commit to it? Tell us what you are planning, and you will have an itemized quote within 2 to 4 business hours, no obligation.
