A nursing logic model makes the reasoning behind a proposed intervention visible. It connects a defined care or service problem with the resources required, the work to be completed, the outputs produced and the outcomes the project expects to influence.
The diagram is useful only when the links are defensible. An arrow from staff training to improved patient outcomes is not evidence of causation; it is a hypothesis about how change may occur and should be supported, measured and revised when implementation produces new information.
What a strong nursing logic model needs
- A specific, evidenced need and clearly defined project boundary.
- Inputs that are genuinely available.
- Activities described as concrete actions.
- Outputs separated from outcomes.
- Short-, intermediate- and longer-term outcomes where relevant.
- Visible assumptions and external factors.
- Indicators that can test priority parts of the pathway.
- Attention to patient experience, equity, governance and unintended effects.
What is a nursing logic model?
A logic model is a concise representation of how a programme or improvement project is expected to work. The U.S. Centers for Disease Control and Prevention describes logic models as tools for connecting programme activities with intended outcomes and for supporting evaluation planning (Centers for Disease Control and Prevention [CDC], 2024a, 2024b).
In nursing, a logic model may support a quality-improvement project, DNP or capstone project, community intervention, education programme, patient-safety initiative or programme evaluation.
The model usually contains:
- Need and context: the problem and conditions surrounding it.
- Inputs: people, time, information, equipment and other resources.
- Activities: what the team will actually do.
- Outputs: the direct products or reach of those activities.
- Outcomes: the changes expected to follow.
- Assumptions/external factors: conditions that may help or hinder the pathway.
Logic model, process map and theory of change
These tools are related but not identical. A process map shows how work currently moves through a system. A theory of change provides a fuller explanation of why and under which conditions an intervention is expected to work. A logic model usually condenses the planned pathway into a more accessible sequence of resources, work and results.
A project may use all three. For example, a process map can identify where discharge communication fails, a theory of change can explain why teach-back might improve understanding, and the logic model can connect the required resources, teaching activity, patient reach, understanding and follow-up outcomes.
1. Define the project boundary
Specify the population, setting, time period and decision the model is intended to support. A six-week test on one adult medical ward requires a different model from an organisation-wide medication-safety programme.
Clear boundaries make the proposed pathway easier to evaluate and reduce the temptation to include every possible activity or outcome.
2. Write an evidence-based need statement
Begin with the problem rather than the preferred intervention. “Medication safety is important” is too broad. A stronger need statement might identify incomplete admission medication histories among eligible patients on a defined unit and explain the associated safety concern.
Use genuine local data where available and explain the denominator, measurement period and limitations. If baseline information has not yet been collected, state that it is required rather than inventing a figure.
3. Identify realistic inputs
Inputs are the resources that make the work possible. Examples include:
- nursing and pharmacy time;
- educator or improvement-lead support;
- electronic-record access;
- patient or carer involvement;
- interpreting or accessibility support;
- training space or digital systems; and
- data collection and analysis capacity.
If a required resource is unavailable, the activity depending on it may not be feasible. The model should reveal that constraint rather than hide it.
4. Describe activities precisely
Activities are what the project team will do. Use concrete verbs such as train, redesign, audit, coach, co-produce, test or review.
“Improve communication” is not specific enough. “Use a teach-back checklist before discharge for eligible patients” describes an observable activity that can be measured.
5. Separate outputs from outcomes
Outputs show what was delivered or who was reached. Outcomes show what changed.
| Element | Example | Interpretation |
|---|---|---|
| Activity | Deliver medication-reconciliation training | What the team does |
| Output | 32 of 40 eligible nurses complete the session | Reach/delivery |
| Short-term outcome | Improved ability to identify reconciliation discrepancies | Change in capability |
| Intermediate outcome | More complete admission reconciliation | Change in care process |
| Longer-term outcome | Fewer preventable medication-related harms | Distant clinical result requiring stronger evidence |
Training attendance is not evidence that competence improved, and improved documentation is not automatically evidence that patient harm decreased. Keeping these levels separate prevents overclaiming.
6. Work backwards from the desired outcome
Identify the meaningful result, then ask what needs to change first. Continue until you reach feasible activities and inputs.
For example, if the desired outcome is fewer unresolved medication discrepancies, reliable reconciliation may need to improve first. That may depend on staff capability, access to medication information and a workflow that makes pharmacist referral straightforward.
Working backwards helps identify missing mechanisms that would otherwise appear as unsupported arrows.
7. Check the forward logic
Read the model from left to right using an “if–then” test:
If this input is available and this activity occurs, what output should follow? If that output occurs, what change becomes plausible, and why?
If the explanation jumps from “one education session” directly to “reduced readmissions,” intermediate mechanisms are probably missing.
8. State assumptions and external factors
Assumptions are conditions the project expects to be reasonably true. External factors are influences outside direct project control.
Examples include staff having protected time to participate, patients being able to engage with the intervention, availability of community records, staff turnover, seasonal workload or changes in organisational policy.
Making these conditions visible improves critical analysis because the reader can see where the pathway may fail.
9. Include patient experience and equity
Ask who could benefit less from the proposed intervention. A digital education programme may work differently for people with limited connectivity, language barriers, sensory impairment or low digital confidence.
Where appropriate, include patients or carers in defining the problem, reviewing materials or interpreting feasibility. Describe their actual contribution rather than labelling a one-off consultation as co-production.
10. Turn the model into an evaluation plan
Priority outputs and outcomes should have practical indicators. Define what will be measured, where the data come from and how the result will be interpreted.
| Model element | Possible indicator | Data source | Important caution |
|---|---|---|---|
| Training output | Proportion of eligible staff completing training | Attendance records | Attendance does not prove competence |
| Process outcome | Eligible admissions with reconciliation completed within 24 hours | Structured record audit | Documentation may not equal accuracy |
| Balancing measure | Median additional nursing time per admission | Workflow/time sample | Early implementation may temporarily take longer |
| Clinical outcome | Clinically important unresolved discrepancies | Defined pharmacist review | Case mix and reviewer decisions may affect rates |
Indicators should answer the evaluation question, not merely use whatever data are easiest to collect.
Worked nursing logic-model example
Consider an adult medical unit with inconsistent medication reconciliation during admission.
| Need/context | Inputs | Activities | Outputs | Expected outcomes |
|---|---|---|---|---|
| Incomplete histories; variable community-record access; peak admission workload | Nurse/pharmacist time; electronic records; educator; patient information; audit support | Co-design a prompt; train staff; test workflow; review discrepancies; provide feedback | Staff trained; eligible admissions using prompt; referrals completed | More complete histories; earlier discrepancy resolution; more reliable reconciliation |
The project should not claim that these activities will automatically reduce patient harm. Instead, it should evaluate whether each nearer-term part of the pathway works and then make longer-term conclusions proportionate to the design and follow-up.
Common nursing logic-model mistakes
- Starting with activities before defining the problem.
- Confusing training delivered with improvement achieved.
- Using vague boxes such as “engagement” or “better care.”
- Drawing long causal jumps with no intermediate mechanism.
- Ignoring staffing, workload, policy or technology context.
- Using a model so detailed that the visual becomes unreadable.
- Claiming causation from a simple before-and-after evaluation.
- Inventing baseline data, stakeholder involvement or project outcomes.
How to write about the model critically
Do not read each box aloud in the dissertation or presentation. Explain the most important relationships, justify the intervention logic with appropriate evidence and identify the assumptions most likely to affect implementation.
A useful analytical sequence is:
Claim → supporting evidence → local application → assumption or limitation → evaluation measure.
This converts the model from a diagram into a defensible project framework.
Academic integrity and governance
The model must describe the real project. Do not invent service data, patient involvement, consultation, implementation activity or results to make the pathway appear complete.
Remove unnecessary patient or staff identifiers from project material and follow the correct organisational route for quality improvement, service evaluation, audit or research. Classification and approval requirements depend on the actual activity and setting.
For broader project standards and service boundaries, see About Nursing Dissertation Service. The student feedback and review policy explains how feedback may be handled and published; it is not presented as a page of verified testimonials.
Logic-model checklist
- The need is specific, evidenced and bounded.
- Every activity depends on an identified input.
- Outputs are direct products rather than claimed benefits.
- Outcomes are observable and separated by time horizon where useful.
- Important arrows can be explained as defensible if–then relationships.
- Assumptions, external factors and equity issues are visible.
- Priority outputs and outcomes have feasible indicators.
- The evaluation includes potential unintended consequences.
- Conclusions will remain proportionate to the evaluation design.
- The visual is readable and supported by concise narrative explanation.
Frequently asked questions
How detailed should a nursing logic model be?
Include enough detail to make the pathway testable while keeping the main visual readable. Operational detail can sit in a companion table or narrative.
Do I need evidence for every arrow?
Prioritise evidence for the most important and uncertain links. Some operational assumptions can come from local knowledge, but they should be explicit and tested where possible.
Can a nursing logic model include patient experience?
Yes. Patient experience may be an outcome, a balancing measure or an input into intervention design. Define how it will be measured or incorporated.
Can the model change during implementation?
Yes. Revision is appropriate when implementation reveals a weak assumption or unexpected pathway. Keep a dated record of meaningful changes.
Does a logic model replace a theoretical framework?
Not necessarily. A logic model organises a programme pathway. A separate theoretical or conceptual framework may still be needed to explain behaviour, implementation or another research phenomenon.
Related Nursing Guides
- Fishbone Diagram Healthcare Example: A Nursing QI Guide
- How to Diagnose and Fix DNP Project Manuscript Problems
- How to Write a Nursing Quality Improvement Project: 12 Steps
Conclusion
A nursing logic model is most useful when it exposes the reasoning behind an intervention. Define the need, distinguish activities from results, identify assumptions, attach meaningful measures and revise the model when evidence changes. The result should be a testable explanation of how the project is expected to work rather than a decorative chain of boxes.
For project-specific help connecting an intervention to implementation and evaluation, use nursing evidence-based practice project support or the contact page.
References
- Centers for Disease Control and Prevention. (2024a, August 18). Step 2 – Describe the program. https://www.cdc.gov/evaluation/php/evaluation-framework-action-guide/step-2-describe-the-program.html
- Centers for Disease Control and Prevention. (2024b, August 20). CDC Program Evaluation Framework. https://www.cdc.gov/evaluation/php/evaluation-framework/index.html
