What is the difference between clinical audit and quality improvement?
Both aim to improve care, and the Healthcare Quality Improvement Partnership (HQIP) describes clinical audit as a quality improvement cycle. The practical difference is in the question each asks and the way each measures.
| Feature | Clinical audit | Quality improvement (QI) |
|---|---|---|
| Core question | Are we doing what we should, against a standard? | How can we make this better, and is our change working? |
| Starting point | An explicit, evidence-based standard with a target | A specific, measurable aim with a date |
| Measurement | Snapshots: a sample before and after change | Frequent, continuous data plotted over time |
| Change | Action plan after the first cycle | Small changes tested in rapid PDSA cycles |
| Finished when | Re-audit shows whether practice improved (closing the loop) | Improvement is sustained and spread |
| Best for | Measuring a gap against clear guidance | Complex problems or where no single standard exists |
How are audit, QI, research and service evaluation different?
| Activity | Question it asks | Example |
|---|---|---|
| Research | What is the best thing to do? Generates new knowledge | Does drug A reduce mortality more than drug B? |
| Clinical audit | Are we meeting the standard? | Do 95% of eligible admissions have a VTE risk assessment within 24 hours? |
| Service evaluation | What does this service achieve now? No preset standard | How long do patients wait for physiotherapy after referral? |
| Quality improvement | Does this change make care better over time? | Can we reduce inpatient falls by testing changes in small cycles? |
Research usually needs research ethics committee approval. Audit, service evaluation and QI normally do not, but they must still be registered and approved through your organisation's audit or QI process. The Health Research Authority publishes a decision tool to help classify projects. Always check with your audit department before collecting any data.
How do I do a clinical audit step by step?
- Choose a topic: Pick something important to patients, common or high-risk, with a clear standard and realistic scope for change. Incidents, complaints and new guidance are good prompts.
- Set the standard, criteria and target: Use NICE, a Royal College, a national alert or an approved trust guideline. Write SMART criteria, a numeric target and defined exceptions.
- Register the audit: Your audit or governance team will check it is an audit, advise on sampling and data protection, and record your role.
- Plan and pilot data collection: Define the population and sample, design a short yes/no/not applicable form, and test it on a few records.
- Collect and analyse: Use audit numbers, not patient identifiers. Calculate compliance as eligible patients meeting the criterion divided by all eligible patients.
- Act: Look for patterns, agree an action plan with owners and timescales, and present at the governance meeting.
- Re-audit: Repeat with the same criterion and method after the changes to close the loop.
Standard: doctors should do VTE assessments properly. We will see how good we are.
Standard: the trust thromboprophylaxis guideline, based on NICE guidance, requires adult inpatients to have a VTE risk assessment on admission. Criterion: adults admitted to the medical admissions unit have a documented assessment within 24 hours of the decision to admit. Target: 95%. Exceptions: discharge within 12 hours; end of life care where thromboprophylaxis is not appropriate.
A worked audit cycle with real numbers
After a patient developed a pulmonary embolism without a documented VTE risk assessment, Wei registers an audit using the criterion above. She reviews 50 consecutive admissions. Four meet the exceptions, leaving 46 eligible patients, of whom 29 had an assessment within 24 hours.
Compliance is 29 divided by 46, multiplied by 100: 63%. Including the excluded patients in the denominator would give 58%, which would be wrong. Among the 17 missed assessments, most were overnight admissions on a paper proforma with no VTE prompt, and junior doctors were unsure who was responsible.
| Measure | First audit | Re-audit (3 months) | Target |
|---|---|---|---|
| Eligible patients | 46 | 47 | n/a |
| Assessment within 24 hours | 29 | 43 | n/a |
| Compliance | 63% | 91% | 95% |
Actions included a mandatory prompt on the proforma, a clear statement that the admitting clinician completes the assessment, short induction teaching and a ward round checklist item. Practice improved but has not yet reached target, so Wei plans to extend the change to specialty wards and re-audit again. Report actual numbers as well as percentages, because small samples can vary by chance.
How do I run a quality improvement project with PDSA cycles?
A widely used framework is the Model for Improvement, promoted by the Institute for Healthcare Improvement. It asks three questions: What are we trying to accomplish? How will we know a change is an improvement? What change can we make that will result in improvement? Changes are then tested in Plan-Do-Study-Act (PDSA) cycles.
- Plan: decide the change, who, where and when, what you predict, and what data you will collect.
- Do: test on a small scale, such as one bay, one shift or one week, and record problems.
- Study: compare the results with your prediction.
- Act: adopt, adapt and retest, or abandon and try something else.
| Measure type | Purpose | Falls project example |
|---|---|---|
| Outcome | Is the system improving for patients? | Inpatient falls per 1,000 bed days |
| Process | Is the change being done? | Percentage with falls risk assessment and lying and standing blood pressure |
| Balancing | Is the change causing problems elsewhere? | Reduced mobility or extra nursing time |
Plot QI data on a run chart with the date along the bottom and each PDSA cycle annotated. Draw a line at the baseline median. Common rules suggest a real change when six or more consecutive points sit on one side of the median, or five or more points rise or fall in a row. One good week is not enough.
Should my project be an audit or QI?
If there is a clear national or local standard, start with an audit to measure the gap. If the problem is complex, has no single standard, or you need to learn which change works, use QI. Often the strongest approach combines them.
We will do a QI project by collecting data for a month, making changes, then collecting data again in six months. If the numbers go up it worked.
Because there is a clear standard, we will audit compliance first. We will then set a specific aim, track a weekly process measure on a run chart, test changes in small PDSA cycles starting with one team, and watch a balancing measure. Once performance is stable, we will re-audit against the standard.
How do I make changes stick and present my project?
- Involve the people who do the work, including nurses, therapists, pharmacists, admin staff and patients where appropriate.
- Find a senior sponsor, such as a consultant or ward manager.
- Prefer system changes (prompts, checklists, defaults) over reminders and memory.
- Share results without blame; the aim is to improve systems.
- Plan the re-audit date and name who will finish the project if you rotate.
Present at your departmental governance or audit meeting in five to ten minutes: background, standard or aim, method, results with numbers and a simple chart, actions and next cycle. Ask the audit team to confirm your role. In applications and interviews, state your role honestly: did you lead it, collect data, or contribute to one part?
I did an audit on VTE and it was very successful.
I led a two-cycle audit of VTE risk assessment against a 95% standard. The first cycle showed 63%, mainly in overnight admissions. With the clinical lead I added a proforma prompt and induction teaching. The re-audit showed 91%. I presented both cycles at governance and we are extending the change.
Where to go next
Our course Clinical Audit and Quality Improvement walks through a full audit cycle with calculations, then the Model for Improvement, PDSA cycles and run charts, with a reflective exercise you can use for appraisal or revalidation. If you want to use your project in job applications, NHS Job Applications and Supporting Statements and NHS Interview Skills show how to present it clearly.
Frequently asked questions
Is a clinical audit a type of quality improvement?
Yes. HQIP describes clinical audit as a quality improvement cycle that measures care against agreed standards and acts to bring practice in line. In everyday NHS use, though, 'QI project' usually means using improvement methods such as the Model for Improvement and PDSA cycles with frequent measurement over time, which is different from the snapshot approach of audit.
Does a clinical audit need ethics approval?
Clinical audit does not need research ethics committee approval, and neither do service evaluation or QI in most cases. Research does. However, audits and QI projects must still be registered and approved through your organisation's clinical audit or governance process, and must follow information governance rules. If unsure how to classify a project, use the Health Research Authority decision tool.
What does closing the audit loop mean?
Closing the loop means completing the audit cycle by making changes after the first data collection and then re-auditing with the same criteria and method to see whether practice has improved. An audit without a re-audit is incomplete. Interview panels often ask whether your audit was completed, meaning whether the loop was closed.
How many patients do I need for a clinical audit?
There is no fixed number. Local audits often use a consecutive sample, such as all eligible patients over two weeks or a fixed number like 50 records. The sample should be realistic and large enough to show a pattern. Your audit department can advise. Report actual numbers alongside percentages, because small samples can change by chance.
What is a PDSA cycle?
PDSA stands for Plan, Do, Study, Act. It is a method for testing a change on a small scale: plan the test and predict the result, carry it out, study the data against your prediction, then act by adopting, adapting or abandoning the change. Several quick cycles, each building on the last, are usually better than one large change.
What is a balancing measure in QI?
A balancing measure checks that your change is not causing problems elsewhere in the system. For example, a falls project that reduces falls by restricting mobility could harm patients in other ways, so you might monitor mobility or staff time. Including a balancing measure shows the panel or governance meeting that you have thought about unintended consequences.
Can I put an audit I only collected data for on my CV?
Yes, as long as you describe your role accurately. Say that you collected data or contributed to a specific part, rather than claiming you led it. Overstating your role is a probity issue and is often exposed by follow-up questions at interview. A smaller role described honestly, with what you learned, is still valuable evidence.
Related courses
Clinical Audit and Quality Improvement
This course teaches you how to plan, complete and present a clinical audit, and how quality improvement differs from audit. It walks through a…
NHS Job Applications and Supporting Statements
This course teaches you how NHS applications are shortlisted and how to write an application and supporting statement that clearly matches the person…
NHS Interview Skills
This course prepares you for NHS interviews by explaining how panels assess candidates and how to give structured, evidence-based answers. It covers…
Sources and further reading
- What is clinical audit and other frequently asked questions. Healthcare Quality Improvement Partnership.
- Best practice in clinical audit. Healthcare Quality Improvement Partnership.
- Defining research table and decision tool. Health Research Authority.
- Venous thromboembolism in over 16s (NG89). National Institute for Health and Care Excellence.
- Quality improvement made simple. The Health Foundation.
- Good Medical Practice (2024). General Medical Council.
Practitioner Insight is an independent education provider. It is not affiliated with or endorsed by the GMC, MPTS, NMC, HCPC, GPhC or any NHS organisation, and this information is not legal advice. If you have been referred to a regulator, speak to your defence organisation, trade union or legal representative.
More guides
How to Write an NHS Supporting Statement (With Examples)
How to write an NHS supporting statement that gets shortlisted: map the person specification, use headings and STAR examples, with weak and strong examples.
Common NHS Interview Questions and How to Answer Them
Common NHS interview questions with example answers: tell us about yourself, why this trust, STAR, clinical, prioritisation, ethics and audit questions.
The GMC Portfolio Pathway (Formerly CESR): A Step-by-Step Guide
The GMC Portfolio pathway (formerly CESR) explained step by step: what changed in 2023, specialty guidance, evidence, referees, timelines and review options.