JHNEBP Model: PET Process & Appraisal Tools

The Johns Hopkins Nursing Evidence-Based Practice (JHNEBP) model is a three-phase process — Practice question, Evidence, Translation (PET) — paired with two purpose-built appraisal tools for scoring the strength and quality of whatever evidence a team finds. That pairing is what distinguishes it from the Iowa Model of Evidence-Based Practice, which it gets confused with constantly: the Iowa Model is a decision pathway of go/no-go organizational gates, while JHNEBP is an appraisal-and-translation method a team works through once it already has a practice question in hand. Different decision structure, different appraisal tools, and answering a different question.

What the JHNEBP Model Is, and Where It Comes From

The model originated at Johns Hopkins Hospital and the Johns Hopkins University School of Nursing, developed by a team of nurse scientists and clinical leaders to give bedside nurses and unit-based teams a structured way to move from a clinical question to a practice change grounded in appraised evidence — not just to describe EBP conceptually, but to operationalize it with tools a working team can actually use on a shift. It has gone through several published editions since it was first formalized, most recently a 2022 edition of the accompanying model-and-guidelines text, and remains one of the most widely adopted nursing EBP frameworks in Magnet®- and Pathway to Excellence®-designated hospitals alongside the Iowa Model.

Where the Iowa Model organizes an entire organizational decision — should we change practice at all, and how do we get from trigger to sustained adoption — the JHNEBP model organizes the appraisal-and-translation work itself. A team can reach JHNEBP’s PET process already knowing it wants to act; what JHNEBP adds is a disciplined way to question, appraise, and translate the evidence once it’s in front of them.

The PET Process: Practice Question, Evidence, Translation

PET is the model’s operating cycle, and each of its three phases carries its own sub-steps rather than being a single undifferentiated stage.

Phase 1 — Practice Question

The team recruits interprofessional members, develops and refines an answerable clinical or operational question (commonly framed with PICO — population, intervention, comparison, outcome — the same framing widely used in evidence-based medicine generally), and defines the scope of the problem before searching for anything. A vague or poorly bounded question at this stage produces an unmanageable evidence search later, so this phase is treated as real work, not a formality on the way to the search.

Phase 2 — Evidence

This is where the two appraisal tools do their work. The team conducts a systematic search of relevant sources, then appraises each piece of evidence individually for its type, level, and quality, summarizes the appraised evidence as a set, and develops recommendations based on the overall strength of what was found. A search that turns up fifteen sources produces fifteen individual appraisals before anyone looks at the evidence as a body — the individual appraisal step is not optional or summarized away.

Phase 3 — Translation

The team determines whether the recommendation is feasible, appropriate, and fits the local setting; creates an action plan; secures approval and implements the change; evaluates outcomes against the plan; reports the results, both internally and externally; and identifies next steps for further inquiry. Translation is deliberately the model’s longest phase in practice, even though it’s the shortest to name — it’s where an appraised recommendation either becomes an actual, sustained change in how a unit works, or doesn’t.

The Evidence Appraisal Tools

JHNEBP’s defining mechanical difference from other EBP models is that it routes every piece of evidence through one of two dedicated appraisal instruments, chosen by evidence type before appraisal begins — not one generic checklist applied uniformly regardless of what’s being appraised.

The Research Evidence Appraisal Tool is used for evidence generated through a formal research process — experimental and quasi-experimental studies, non-experimental studies, qualitative studies, and systematic reviews or meta-analyses of any of those. It guides the appraiser through study design, sample and setting, measurement validity and reliability, and results, ending in a Level and a Quality grade for that individual source.
The Non-Research Evidence Appraisal Tool is used for evidence that didn’t come from a formal research study but still bears on the practice question — clinical practice guidelines and consensus panel statements, literature and integrative reviews, organizational experience data (quality improvement or program-evaluation findings, financial or benchmarking data), case reports, and the opinion of recognized experts. It asks different questions than the research tool — is the guideline current and its recommendations clearly graded, is the QI data methodologically sound for its own limited purpose — because a case report and an RCT fail in different ways and need different scrutiny.

Routing evidence to the correct tool before appraisal, rather than forcing every source through one instrument, is precisely the mechanical distinction this guide’s opening paragraph means by “different appraisal tools” — the Iowa Model directs a team to critically appraise and synthesize evidence as a single step without prescribing which of two instruments to use for which evidence type.

Evidence Levels and Quality Grades, With a Worked Example

Both appraisal tools resolve to the same two-part rating for each source: a Level (I through V, describing what kind of evidence it is) and a Quality grade (A, B, or C, describing how well that particular piece of evidence was conducted). The levels, from strongest to weakest evidence type:

Level I — experimental study (RCT), or a systematic review of RCTs, with or without meta-analysis.
Level II — quasi-experimental study, or a systematic review of a combination of RCTs and quasi-experimental studies.
Level III — non-experimental study, or a systematic review of a combination of RCTs, quasi-experimental, and non-experimental studies, or of qualitative studies with or without meta-synthesis.
Level IV — opinion of respected authorities and/or nationally recognized expert committees or consensus panels based on scientific evidence, including clinical practice guidelines and consensus panels.
Level V — evidence from literature reviews, quality improvement or program-evaluation data, financial evaluation, case reports, or the opinion of nationally recognized experts based on their experience, without an accompanying critical appraisal.

Quality grades then describe execution within that level: A (High quality) — consistent, generalizable results, sufficient sample, adequate control, definitive conclusions, consistent recommendations from a comprehensive literature review; B (Good quality) — reasonably consistent results, sufficient sample for the study design, some control, fairly definitive conclusions; C (Low quality or major flaws) — little evidence of control, inconsistent or insufficient sample, conclusions cannot be drawn.

A worked example. Say a unit’s practice question, developed in the Practice Question phase, concerns a fall-prevention bundle. The team’s search returns, among other sources, a single-site quasi-experimental pre/post study of the bundle with a defined comparison period, adequate sample size, and consistent, plausible results, but no randomization or control group. Because it’s a formal research study, it goes through the Research Evidence Appraisal Tool, not the non-research tool. Its design — quasi-experimental, not randomized — places it at Level II. Because its results are reasonably consistent and the sample and design are adequate for that type of study, but it lacks the tighter control an RCT would have, it’s graded Quality B. That Level II/Quality B rating is then weighed alongside every other appraised source — a Level IV consensus guideline on fall prevention appraised through the non-research tool, say, or a Level V internal QI report from a comparable unit — when the team synthesizes an overall recommendation strength in the rest of the Evidence phase. No single source’s rating stands alone; it’s an input to synthesis, not a verdict.

How JHNEBP Differs From the Iowa Model

Both are established, widely taught nursing EBP frameworks and both show up on the same reading lists, but they answer different questions and hand a team different tools:

What each one is. The Iowa Model is an organizational decision pathway with explicit go/no-go gates and two feedback loops — it decides whether a trigger becomes a sustained practice change at all. JHNEBP is an appraisal-and-translation method — the PET process — for a team that already has a practice question and needs a disciplined way to search, appraise, and translate the evidence behind it.
How evidence gets appraised. The Iowa Model asks a team to critically appraise and synthesize evidence as a single combined step, without prescribing separate instruments by evidence type. JHNEBP routes every source through one of two dedicated instruments — the Research Evidence Appraisal Tool or the Non-Research Evidence Appraisal Tool — chosen before appraisal begins, each resolving to an explicit Level (I–V) and Quality (A–C) rating.
Where the branching happens. The Iowa Model’s branching lives in its decision points — is this a priority, is there sufficient evidence, is the change appropriate for adoption — each with an explicit loop-back path. JHNEBP’s structure is sequential across its three PET phases rather than gated; its rigor sits inside the Evidence phase’s appraisal step rather than in stage-gate decision points.
What a program gains from each. A program adopting the Iowa Model gets a governance-level framework for deciding what gets pursued and sustained organization-wide. A program adopting JHNEBP gets a repeatable, taught method — often paired with unit-level EBP mentors — for how any given question gets worked once someone decides to pursue it. The two are complementary rather than exclusive: an organization can use the Iowa Model to decide whether and when to act, and JHNEBP’s PET process as the actual working method a unit team follows once it does.

When the JHNEBP Model Fits

JHNEBP fits well when a unit or department already has a specific, answerable practice question and needs a structured, teachable method for a team — often including staff nurses without prior EBP training — to search, appraise, and translate evidence into a local change. Its explicit Level/Quality rating system is also useful independent of the full PET cycle: many programs adopt the appraisal tools alone as a shared vocabulary for grading evidence strength in journal clubs, unit councils, or other EBP frameworks’ evidence-appraisal steps. It fits less well as a tool for deciding, at an organizational level, whether a topic is worth pursuing in the first place, or for allocating implementation capacity across competing priorities — that decision-gate work is what the Iowa Model, not JHNEBP, is built for. It’s also not a substitute for formal systematic-review methodology: appraising evidence for a single unit’s practice question with these tools is a lighter-weight exercise than a registered systematic review with its own protocol, which belongs to evidence-synthesis methodology rather than this model.

Frequently Asked Questions

What does PET stand for in the Johns Hopkins Nursing EBP model?

Practice question, Evidence, Translation — the three sequential phases of the model. Practice Question forms the team and the answerable question; Evidence searches, appraises, and synthesizes the evidence; Translation determines fit, implements the change, and evaluates and disseminates the outcome.

What’s the difference between the Research and Non-Research Evidence Appraisal Tool?

The Research Evidence Appraisal Tool is used for evidence from a formal research process — experimental, quasi-experimental, and non-experimental studies, qualitative studies, and systematic reviews or meta-analyses. The Non-Research Evidence Appraisal Tool is used for evidence that didn’t come from a research study but still bears on the practice question — clinical practice guidelines, literature reviews, organizational quality-improvement or financial data, case reports, and expert opinion. Which tool applies is decided by evidence type before appraisal starts.

How many evidence levels does the JHNEBP model use, and what do they mean?

Five, Level I through Level V, ranked by evidence type from strongest to weakest: Level I is experimental research (RCTs) or systematic reviews of RCTs; Level II is quasi-experimental research; Level III is non-experimental or qualitative research; Level IV is expert/consensus-panel opinion including clinical practice guidelines; Level V is literature reviews, QI or program-evaluation data, case reports, or individual expert opinion without formal critical appraisal. Each source also gets a separate Quality grade (A, B, or C) describing how well it was conducted within its level.

Is the JHNEBP model the same as the Iowa Model?

No, despite frequently appearing on the same reading lists. The Iowa Model is an organizational decision pathway with go/no-go gates deciding whether a trigger becomes a sustained practice change. JHNEBP is an appraisal-and-translation method (the PET process) for a team that already has a practice question, built around its own two dedicated evidence-appraisal instruments. See the full comparison above.

Can the JHNEBP appraisal tools be used without the full PET process?

In practice, yes — many programs use the Research and Non-Research Evidence Appraisal Tools, and the Level/Quality rating system they produce, as a standalone shared vocabulary for grading evidence strength (in a journal club or unit EBP council, for example) independent of running a full Practice Question–Evidence–Translation cycle.

Johns Hopkins Nursing EBP Model: The PET Process and Evidence Appraisal Tools

Ask about Johns Hopkins Nursing EBP Model: The PET Process and Evidence Appraisal Tools