Ask a nurse who trained on the Johns Hopkins Nursing Evidence-Based Practice Model to name its three phases, and most will answer instantly: Practice question, Evidence, Translation. That three-letter shorthand, PET, is one of the reasons the model has become one of the most widely taught EBP frameworks in nursing programs, alongside the Iowa Model and the ACE Star Model.
What sets the Johns Hopkins Model apart from several of its peers is not the three-phase structure itself, several other frameworks organize work in roughly the same order, but the detailed evidence-rating machinery built into the Evidence phase: a Level I through V strength scale and an A through C quality scale applied to every individual source. For a capstone committee that wants to see exactly how you weighed your evidence, that built-in rating tool does a lot of the justification work for you.
This guide walks through where the model came from, what each PET phase actually requires, the evidence rating tools in full, and a complete worked example at capstone scale. For the wider landscape of EBP frameworks this model sits alongside, see our nursing theory and change frameworks hub. If your program also expects a general evidence-based practice discussion beyond a single named model, our evidence-based practice guide covers that broader ground.
What Is the Johns Hopkins EBP Model?
The Johns Hopkins Nursing Evidence-Based Practice Model was developed by a team of nurse leaders at Johns Hopkins Hospital and the Johns Hopkins School of Nursing in the early 2000s, built specifically to give practicing nurses and nursing students a structured, teachable process for moving from a clinical question to an evidence-based practice change. It has been refined and republished across several subsequent editions since its original release, with updated appraisal tools and expanded guidance each time.
Unlike frameworks built primarily as academic or theoretical models, the Johns Hopkins Model was designed from the start as a practical, hospital-based tool, complete with standardized forms, a project management guide, and the appraisal tools that make it easy to teach consistently across a large nursing workforce. That practical, form-driven design is part of why so many nursing programs adopt it as their capstone framework of choice: students get concrete templates rather than an abstract description of a process.
The model organizes work into three broad phases, Practice Question, Evidence, and Translation, each containing several concrete steps. Unlike the Stetler Model's individual-practitioner framing, the Johns Hopkins Model assumes an interprofessional team from the very first step, which makes it a natural fit for a capstone project built around a unit-level or organizational practice concern rather than one nurse's personal decision.
The Three Phases of the Johns Hopkins EBP Model (PET)
| Phase | What it covers | Typical output |
|---|---|---|
| Practice Question | Recruit an interprofessional team, develop and refine an answerable question, define the project's scope, and secure leadership support | A clearly scoped, answerable practice question with a sponsoring team |
| Evidence | Search internal and external sources, appraise each one, rate its strength and quality, summarize the evidence, and develop a recommendation | A rated evidence table and a stated recommendation |
| Translation | Determine fit, feasibility, and appropriateness, build an action plan, implement the change, evaluate outcomes, and report and disseminate results | An implemented, evaluated practice change with a dissemination plan |
Notice that the Evidence phase is where this model asks for more structured work than most of its peers. Where the Iowa Model or PARIHS discuss appraisal in general terms, the Johns Hopkins Model hands you a specific rating scale to apply to every source, which is exactly what most capstone rubrics are looking for in an appendix table.
Applying the Practice Question Phase
The Practice Question phase is where the project's foundation gets built, and where a rushed capstone most often runs into trouble later. Four concrete steps make up this phase.
Recruit an Interprofessional Team
Identify the people who touch the practice concern directly, not just those most convenient to ask. At capstone scale this is often small: yourself, a unit-level champion or preceptor, and one or two stakeholders affected by the eventual change, such as a charge nurse or a unit educator.
Develop and Refine the Practice Question
State the clinical concern as a specific, answerable question rather than a general topic. Many students frame this using a PICO-style structure, naming the population, the intervention or change under consideration, a comparison where relevant, and the outcome the project is meant to affect.
Define the Scope of the Practice Question
Decide explicitly whether the question applies to a single unit, a department, or an entire facility, and be honest about what your timeline and access actually support. A question scoped too broadly for a single-semester capstone is one of the most common reasons projects stall midway through.
Secure Support and Determine Responsibility for Leadership
Confirm who is accountable for moving the project forward and secure whatever sign-off your setting genuinely requires, a unit manager's informal approval for a small-scale change, or a formal committee review for anything touching policy, staffing, or patient safety protocols.
Applying the Evidence Phase
The Evidence phase is the model's most distinctive stretch of work, and the one that generates the appraisal table most capstone committees expect to see in an appendix. Five steps make up this phase.
Conduct an Internal and External Search for Evidence
Search published literature through your library's databases, and also check internal sources: existing unit policies, prior quality-improvement data, and any relevant guidelines your organization has already adopted. Internal evidence often gets overlooked, even though it can be some of the most directly applicable evidence available.
Appraise the Level and Quality of Each Source
Read each source critically for its research design, sample, and methodological rigor, then assign it a strength level and a quality grade using the model's rating tools, covered in full further down this page. This step is where the model's evidence table takes shape.
Summarize the Individual Sources
Write a brief, factual summary of what each rated source actually found, not just its topic. A useful summary states the population studied, the intervention, and the specific result, in language specific enough that a reader could tell your sources apart without rereading the original articles.
Synthesize the Overall Strength of the Evidence
Step back from the individual sources and describe what they collectively show: whether findings are consistent across studies, how strong the highest-quality sources are, and where the evidence base is thin, mixed, or silent on part of your question.
Develop a Recommendation
State plainly what the synthesized evidence supports: proceeding with a specific practice change, proceeding cautiously with close monitoring, or gathering more evidence before acting. This recommendation is the explicit bridge into the Translation phase, so avoid leaving it implied.
Building your rated evidence table?
Our nursing writers can take your specific practice question and the sources you have gathered, and build out a fully rated Level and Quality evidence table with the synthesis and recommendation the Evidence phase requires.
Get DNP project help →Applying the Translation Phase
Translation is where the recommendation from the Evidence phase becomes an actual, evaluated practice change. Five steps make up this final phase.
Determine Fit, Feasibility, and Appropriateness
Weigh the recommendation against your setting's readiness: available resources, organizational priorities, and whether the timing genuinely supports a change right now. A recommendation can be well supported by evidence and still need to wait for a more feasible moment.
Create an Action Plan
Lay out the specific steps, timeline, and responsible parties for implementing the change. A concrete action plan names who does what by when, rather than describing the change only in general terms.
Secure Support and Implement the Change
Confirm any remaining approvals, then carry out the plan, tracking what actually happens against what was planned. Note any adjustments made along the way, since capstone write-ups that pretend implementation went exactly as scripted tend to read as less credible, not more.
Evaluate Outcomes
Measure the specific outcome named in your original practice question, comparing post-implementation data against a defined baseline. State clearly whether the outcome moved in the intended direction and by how much.
Report the Results and Identify Next Steps, Disseminate Findings
Report what happened honestly, including any gap between expected and actual results, and identify what should happen next: sustaining the change, scaling it further, or revising the approach. Dissemination, a poster, a unit presentation, or a manuscript, closes the loop the model expects every project to reach.
How the Model Rates Evidence Strength and Quality
The rating tool most students associate with this model has two parts: a strength level based on research design, and a separate quality grade based on how well that particular study was conducted. Every source gets both a Level and a Quality rating, not just one.
| Level | Type of evidence |
|---|---|
| Level I | Experimental studies, randomized controlled trials, or systematic reviews of randomized controlled trials |
| Level II | Quasi-experimental studies or systematic reviews combining experimental and quasi-experimental designs |
| Level III | Non-experimental studies, qualitative studies, or qualitative meta-syntheses |
| Level IV | Clinical practice guidelines and consensus panels based on expert opinion |
| Level V | Literature reviews, quality improvement or program evaluation data, financial evaluation, and case reports |
| Quality grade | What it signals |
|---|---|
| A: High quality | Consistent, generalizable results; sufficient sample size; adequate design and controls; definitive conclusions |
| B: Good quality | Reasonably consistent results; adequate sample size for the study design; some limitations, with fairly definitive conclusions |
| C: Low quality or major flaws | Little to no evidence of internal validity or design controls; inconsistent results; conclusions cannot be drawn reliably |
A source's rating comes from combining the two, for example a Level I, Quality A randomized controlled trial carries far more weight in your recommendation than a Level V, Quality C case report. Most capstone appendices present this as a single table: source citation, Level, Quality, and a one or two sentence summary of the finding.
Why Choose the Johns Hopkins Model for Your Capstone
Reach for the Johns Hopkins Model when your program or committee specifically wants to see a rated, source-by-source evidence table, since few other frameworks build that requirement in as directly. It is also a strong fit when your project already has, or can assemble, a genuine interprofessional team from the outset, rather than starting as one nurse's individual observation.
The model's practical, form-driven design also makes it approachable for a first EBP project: the PET labels, the standardized appraisal tools, and the step-by-step structure give a student a clear scaffold to follow rather than an abstract theory to interpret. Many programs choose it precisely because it is easy to teach consistently across a whole cohort.
If your project's real story is one practitioner's individual decision without a team behind it, the Stetler Model will likely carry that framing more naturally. If your committee wants heavier emphasis on organizational context and facilitation as distinct dimensions, PARIHS or i-PARIHS addresses that more directly than this model's fit and feasibility step alone.
A Worked Example: Reducing Unnecessary Urinary Catheter Days
To make each phase concrete, here is how the Johns Hopkins Model plays out on a realistic capstone-scale problem: a medical-surgical unit with indwelling urinary catheters left in place longer than clinically necessary, raising catheter-associated urinary tract infection risk.
Practice Question. The student recruits a small interprofessional team: a charge nurse, the unit's infection-prevention liaison, and a physician assistant who writes most of the unit's catheter orders. Together they frame the question: among adult med-surg patients with an indwelling catheter, does a nurse-driven removal protocol with daily necessity criteria, compared with physician-initiated removal alone, reduce average catheter days? They scope the project to one 24-bed unit over an eight-week implementation window and secure the nurse manager's support.
Evidence. The team searches internal infection-prevention data alongside published literature, gathering eleven sources. Three randomized controlled trials rate Level I, Quality A or B and consistently support nurse-driven removal protocols; four quasi-experimental unit-level studies rate Level II, mostly Quality B; two clinical practice guidelines rate Level IV; and two internal quality reports rate Level V. The synthesis concludes the evidence is strong and consistent for adult med-surg populations, and the team's recommendation is to proceed with implementation.
Translation. Fit and feasibility check out well: the EHR already supports a daily necessity-criteria prompt, and the physician assistant is willing to pre-authorize nurse-driven removal within defined criteria. The action plan names the criteria, a two-week staff education rollout, and a designated go-live date. After implementation, average catheter days drop from a baseline of 4.1 days to 2.6 days over the evaluation period, and the team reports the result at a unit safety huddle with a plan to sustain the protocol and monitor it quarterly.
Notice how each phase produces a specific, checkable output, a scoped question, a rated evidence table, a stated recommendation, and a measured before-and-after result, rather than a general narrative about wanting to improve catheter care. That specificity is exactly what a rubric built around this model is checking for.
The Johns Hopkins Model vs. Other EBP Frameworks
| Model | How it differs from the Johns Hopkins Model |
|---|---|
| Iowa Model | Also assumes an organizational team, but leaves appraisal more general rather than supplying a matching Level and Quality rating tool. See our full Iowa Model guide. |
| PARIHS / i-PARIHS | Treats organizational context and facilitation as separate, explicitly weighted dimensions, where the Johns Hopkins Model folds similar considerations into the Translation phase's fit and feasibility step. See our full PARIHS and i-PARIHS guide. |
| ACE Star Model | Maps the full upstream path evidence travels from primary research through translation, a broader knowledge-transformation lens than this model's project-focused PET sequence. See our full ACE Star Model guide. |
| Stetler Model | Built for an individual practitioner deciding alone, without assuming an interprofessional team from the first step the way the Johns Hopkins Model does. See our full Stetler Model guide. |
If a committee member asks why you chose the Johns Hopkins Model over an alternative, a grounded answer usually points to the evidence rating table: this model gave you a built-in, teachable way to show exactly how strong and how well-conducted each source was, rather than describing your appraisal process in general terms.
Where the Johns Hopkins Model Fits in Your Capstone Document
| Capstone chapter | PET phase it corresponds to |
|---|---|
| Introduction / Problem Statement | Practice Question: the clinical concern, its scope, and the team behind it |
| Literature Review | Evidence: your search strategy, individual source appraisals, and rated evidence table |
| Theoretical/Conceptual Framework | Full explanation of the Johns Hopkins Model and why an interprofessional, evidence-rated approach fits your project |
| Methodology (implementation plan) | Translation: fit and feasibility findings, the action plan, and implementation details |
| Results / Discussion | Translation: outcome evaluation data and what it suggests for sustaining or scaling the change |
| Dissemination | Translation: reporting results and next steps, including your presentation or manuscript plan |
See our DNP evidence synthesis guide for a closer look at building the kind of rated, synthesized evidence table this model's Evidence phase expects.
Common Mistakes When Applying the Johns Hopkins Model
- Naming a Level for a source but skipping the separate Quality grade, or the reverse, when the model expects both
- Framing the Practice Question phase as a solo decision, without naming the interprofessional team the model assumes from the start
- Writing an evidence summary that restates each source's topic without stating what it actually found
- Leaving the recommendation implied at the end of the Evidence phase instead of stating it as an explicit next step
- Describing implementation as if it went exactly according to the action plan, without noting any real adjustments made along the way
- Skipping the dissemination step entirely, treating outcome evaluation as the project's natural endpoint
Need Your Johns Hopkins EBP Model Section Written and Applied?
Our nursing writers can take your specific practice question, build a fully rated evidence table, and write out the Translation section, action plan and all, for your BSN, MSN, or DNP project.
Get nursing writing helpExplore nursing servicesRelated Guides
Frequently Asked Questions
PET stands for Practice question, Evidence, and Translation, the three phases the model organizes work into. Practice Question defines the clinical problem and forms an answerable question; Evidence covers searching, appraising, and rating the literature; Translation covers deciding whether to act on the evidence, building an action plan, implementing it, and evaluating the outcome. Most capstone write-ups use the PET labels directly as section headers, since committees trained in the model recognize them immediately.
Both assume an interprofessional team and an organizational-scale change, so they overlap more than most EBP models do. The clearest difference is the evidence-handling machinery: the Johns Hopkins Model supplies a detailed Level I through V strength scale plus an A through C quality rating for every individual source, while the Iowa Model discusses appraisal in more general terms without a matching built-in rating tool. Choose the Johns Hopkins Model when your program specifically wants that individual-source rating table in your appendix.
For a capstone-scale project, most programs expect the practice, most students actually get value from applying the full Level I through V and Quality A through C scale to the core studies driving their practice recommendation, roughly eight to fifteen sources, rather than every article a search returned. Sources used only for background context in your introduction typically do not need a formal rating. Confirm your own program's expectation, since some rubrics do want every cited source rated.
That is common for emerging or narrowly specialized practice questions, and the model does not require Level I evidence to proceed. The Evidence phase asks you to synthesize what you have honestly, name the strength and quality of each source, and let the overall evidence summary reflect that mix rather than overstating it. A Translation decision built on weaker evidence, cautious, small-scale, closely monitored, is often the more defensible choice than treating thin evidence as if it were strong.
The model scales down cleanly. A BSN or MSN student can run all three phases at a single-unit scope with a smaller interprofessional team, perhaps a preceptor, a unit educator, and a charge nurse, rather than the full committee-level team a DNP project typically assembles. The Practice Question, Evidence, and Translation structure and the evidence rating tools work identically at either scale.
The model does not set a fixed number, and capstone-scale teams are typically small, often just two to four people who touch the problem directly: the student, a unit-level champion or preceptor, and one or two stakeholders affected by the eventual change. What matters more than headcount is that the question genuinely reflects more than one person's individual perspective before it moves into the Evidence phase.
The evidence summary describes what the literature collectively shows: how many sources support a given approach, how strong and consistent that support is, and where the evidence conflicts or falls silent. The recommendation is the separate, explicit next step that follows from that summary, whether to proceed with a specific practice change, proceed cautiously with monitoring, or gather more evidence first. Committees frequently dock points when a paper blends the two into one vague paragraph instead of stating each clearly.
Yes. Our nursing writers can take your specific practice question, build out a rated evidence table using the Level I through V and Quality A through C scale, and write a fully applied Translation section with a concrete action plan and evaluation approach for your BSN, MSN, or DNP project.