Quality Improvement, Clinical Audit and Research in Healthcare
Foundations of quality improvement and quality management
NHS England quality management systems place improvement alongside planning, control and assurance. This module introduces improvement as disciplined system learning rather than isolated project activity.
Learning objectives
• Explain quality as a property of systems rather than a score owned by one profession. • Distinguish quality planning, assurance, control and improvement and recognise where audit fits. • Use the Model for Improvement as a practical learning framework without treating it as the only legitimate method. • Recognise why interprofessional and patient/service-user involvement changes the quality of improvement work. • Identify common failure modes in improvement projects before collecting data.
Learning objectives
- Explain quality as a property of systems rather than a score owned by one profession.
- Distinguish quality planning, assurance, control and improvement and recognise where audit fits.
- Use the Model for Improvement as a practical learning framework without treating it as the only legitimate method.
- Recognise why interprofessional and patient/service-user involvement changes the quality of improvement work.
- Identify common failure modes in improvement projects before collecting data.
Why improvement is not simply “doing better”
Improvement is deliberate, systematic work to change a process and learn whether that change produces better outcomes. A service can be staffed by committed professionals and still produce unreliable care because the system contains delays, hand-off failures, unclear ownership, poor feedback loops or competing priorities. The practical implication is important: when performance is unreliable, asking individuals to “try harder” may temporarily improve behaviour but often does not redesign the conditions that produced the problem.
The quality management view
A useful quality-management model distinguishes planning, assurance, control and improvement. Planning defines what good should look like and how a service intends to deliver it. Assurance asks whether required standards and governance arrangements are in place. Control monitors whether a process remains within expected performance. Improvement changes the system when current performance is not good enough. Clinical audit can contribute to assurance and improvement when it compares care with explicit standards and leads to action and re-audit.
The Model for Improvement
The Model for Improvement asks three linked questions: what are we trying to accomplish, how will we know that a change is an improvement, and what changes can we make that will result in improvement? PDSA cycles are then used to test changes. The model is intentionally simple; the sophistication comes from the quality of the aim, measures, change theory and learning. A PDSA is not a substitute for governance, evidence review or understanding the current system.
Interprofessional improvement
Care pathways cross professional and organisational boundaries. A medicines discharge project may involve prescribers, pharmacists, nurses, ward clerks, community teams, patients and carers. A falls project may involve nursing, therapy, medical staff, estates and equipment services. Building the team around the process rather than the hierarchy reveals hand-offs and constraints that a single professional group may not see.
Patients, service users and carers as partners
Involvement should be purposeful. People who use a service can identify burdens, access barriers and outcomes that routine data miss. They can help define what “better” means, test whether a proposed process is acceptable, and identify unintended consequences. Involvement must be accessible and proportionate, with clarity about whether people are co-designing, advising, providing data or participating in research.
Ethics does not disappear in QI
QI, audit and service evaluation are often lower-risk than research, but they still create ethical obligations. The 2025 HQIP ethics guidance emphasises benefit-risk balance, scientific validity, equity, value and conflicts of obligation. Poorly designed improvement can waste staff time, expose confidential information, burden patients or widen inequalities. Ethical thinking belongs at project design, not as a final administrative check.
Worked example: delayed pain relief
Problem signal: patients on an acute ward report long waits for prescribed analgesia. Weak response: send an email reminding nurses to give medicines promptly. Stronger response: map the medication pathway, define the delay precisely, measure time from clinically appropriate request to administration, identify where delays occur, involve nursing/pharmacy/prescribing staff and patients, and test small changes while monitoring pain outcomes and unintended consequences such as omitted safety checks.
Practical checklist before starting
Before collecting data, be able to state the problem, who experiences it, the setting, the current evidence, the intended project type, the decision the data will support, the people who need to be involved, the likely risks, and who will own the process after the project. If these cannot be answered, data collection is probably premature.
