Evaluating a health programme: structure, process and outcome; formative and summative evaluation
- Safe practice
Evaluating a programme shows whether it was delivered as planned and whether it changed anything, so the team can improve or stop it.
What it is
Programme evaluation collects credible evidence to judge how a programme works and what it achieves. CDC's 2024 framework has six steps, from assessing the context to acting on the findings. Formative evaluation checks, before or during roll-out, whether a programme is feasible and acceptable so it can be improved; summative evaluation judges its effectiveness once delivered, against the planned outcomes. Process evaluation compares delivery with the plan, outcome evaluation measures how far the intended outcomes were reached and impact evaluation estimates what would have happened without the programme. Donabedian's model sorts indicators into structure (resources and context), process (what staff do) and outcome (results of care).
What it requires
- CDC's five evaluation standards: relevance and utility, rigour, independence and objectivity, transparency, and ethics.
- Interest holders include the people served, those who run the programme, those who will use the findings and those who are sceptical of it.
- Outcome evaluation alone cannot show what caused the outcomes.
- Economic evaluation weighs effects against costs (cost analysis, cost-benefit, cost-utility).
- Using several data sources and methods makes the findings more rigorous.
How it is applied (and why)
- Describe the context and involve the interest holders, including the community — the evaluation answers their real questions.
- Describe the programme with its logic model — you cannot judge what you have not defined.
- Choose the questions and the type: formative, process, outcome or impact — the question decides the design.
- Choose structure, process and outcome indicators with exact numerators and denominators — vague indicators give vague answers.
- Gather data from several sources: records, surveys, interviews, observation — different sources check one another.
- Analyse against the objectives and their deadlines — success is judged against what was planned.
- Interpret with the team and the community — people who live the programme explain the numbers.
- Act on the findings: adjust, expand or stop, and report back — an unused evaluation changes nothing.
Types of evaluation applied to a fictional hypertension programme
| Type | Question | Example of evidence |
|---|---|---|
| Formative | Is the plan feasible and acceptable to the community? | Pilot sessions and interviews with leaders before roll-out |
| Process | Was it delivered as planned? | Proportion of planned home visits actually made |
| Outcome | Were the intended outcomes reached? | Proportion of referred adults who attended a control visit |
| Impact | Did the programme cause the change? | Comparison with a similar micro-territory without the programme |
In nursing practice
- Structure example: trained staff and working equipment; process: blood pressure measured correctly; outcome: blood pressure controlled.
- If outcomes fail, check the process first: the programme may not have been delivered as planned.
- Share results with the community and in accountability spaces, not only with the funder.
- Record activities accurately; evaluation is only as good as the records it uses.
Common mistakes
- Evaluating only activities (number of sessions) and calling it impact.
- Claiming the programme caused a change without a comparison.
- Deciding the indicators after the data were collected.
- Writing a report that no one uses to decide anything.
Test yourself
Think of the answer, then tap the question to see it.
What is the difference between formative and summative evaluation?
Formative evaluation improves the programme before or during roll-out; summative evaluation judges its effectiveness once delivered, against the planned outcomes.
In Donabedian's model, is «percentage of visits where blood pressure was measured correctly» structure, process or outcome?
Process: it describes what staff actually did.
Can an outcome evaluation alone show that the programme caused the result?
No. Showing cause needs an impact evaluation that compares with what would have happened without the programme.
What is the last step of CDC's 2024 evaluation framework?
Act on the findings: use them to adjust, expand or stop the programme and report back.
Elsewhere in DurgiPedia
Connections
Educational content for students; it does not replace your institution's protocols, supervised clinical training or a prescriber's order. Dose-calculation examples use made-up numbers: they are educational, not a clinical order.
Sources: CDC Approach to Program Evaluation (types of evaluation) (Centers for Disease Control and Prevention, 2024); CDC Program Evaluation Framework, 2024 (MMWR Recommendations and Reports 73(6)) (Centers for Disease Control and Prevention, 2024); Factors associated with idle peripheral intravenous catheters: a point-prevalence study using the Donabedian model (Joshi P et al.) (BMC Research Notes (PubMed Central), 2026); Formative Evaluation of an Early Family-Centred Prevention Programme for Childhood Overweight and Obesity (fruehstArt): A Study Protocol (Ruettger K et al.) (Children (PubMed Central), 2025).
Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake
Representative model, for educational purposes only.