Quality Management, Patient Safety and Clinical Governance

Introduction and context

Quality in healthcare is the degree to which services increase the likelihood of desired health outcomes and are consistent with current professional knowledge. It is broader than technical competence. It includes access, timeliness, continuity, safety, effectiveness, efficiency, equity, dignity, communication and patient experience.

Healthcare quality is difficult to manage because patients have different needs, outcomes may appear long after care is delivered, and many results depend on patients, families, clinicians, technology and the wider social environment. Quality management therefore requires both professional judgement and systematic organisational methods. A hospital manager must create conditions in which safe care is reliably delivered, problems are detected early, learning is shared and improvement is sustained.

Patient safety is concerned with preventing avoidable harm and reducing unnecessary risk associated with healthcare. Clinical governance is the integrated system through which a healthcare organisation is accountable for continuously improving the quality and safety of care while maintaining high standards of clinical performance.

Learning objectives

By the end of this chapter, the learner should be able to:

1. Explain quality management, patient safety and clinical governance.
2. Apply Donabedian’s structure–process–outcome model.
3. Use continuous quality improvement methods, indicators and clinical audit.
4. Explain accreditation and standards-based assurance.
5. Manage incident reporting, root-cause analysis and Failure Mode and Effects Analysis.
6. Recommend effective infection-prevention and medication-safety practices.
7. Evaluate ethical, managerial and technological implications of quality and safety programmes.

1. Foundations of healthcare quality

Dimensions of quality

The following dimensions are widely used in healthcare quality assessment:

Quality assurance traditionally emphasises compliance with predetermined standards. Quality improvement is more developmental: it studies processes, tests changes and uses data to improve performance. Quality control detects deviations, whereas quality management combines planning, assurance, control and improvement across the organisation.

Clinical governance

Clinical governance is not a single committee or document. It is an organisational framework including:

• clinical effectiveness and evidence-based practice;
• professional education, competence and appraisal;
• clinical audit and quality improvement;
• risk management and patient safety;
• incident reporting and learning;
• patient and public involvement;
• information governance and reliable documentation;
• workforce wellbeing and appropriate accountability.

Effective governance requires clear responsibility. The governing body sets expectations and allocates resources; senior executives create systems and monitor performance; clinical leaders translate standards into practice; managers coordinate data, staffing and improvement; and every employee has a duty to report hazards and act within competence.

2. Donabedian’s structure–process–outcome model

Avedis Donabedian proposed that quality can be examined through three linked domains.

Structure refers to the setting in which care occurs: buildings, equipment, staffing, qualifications, policies, information systems and organisational arrangements.

Process refers to what is actually done: assessment, diagnosis, treatment, communication, hand hygiene, documentation, referral and discharge planning.

Outcome refers to the effects of care on patients and populations: mortality, complications, functional status, symptom relief, readmission, satisfaction and quality of life.

Figure: Structure influences processes, processes influence outcomes, and outcome information feeds organisational learning.

The model is useful but not deterministic. Good structure does not guarantee good care, and poor outcomes may result from disease severity rather than poor treatment. Consequently, indicators should be interpreted in context, with attention to case mix, measurement quality and unintended effects.

Indicators

A quality indicator is a defined, measurable signal of performance. Common types include:

• Structure indicators: proportion of nurses with required competency certification.
• Process indicators: percentage of eligible patients receiving venous-thromboembolism assessment.
• Outcome indicators: postoperative infection rate or mortality.
• Balancing indicators: measures that detect harm caused by an improvement, such as staff workload or emergency-department waiting time after accelerating another service.

A rate is often calculated as:

Here, the numerator is the number of events, the denominator is the population genuinely at risk during a stated period, and K is a multiplier such as 100, 1,000 or 100,000. The numerator must be included in the denominator where appropriate. Rates should include a time period and clear case definitions.

For example, if 8 catheter-associated infections occur during 2,000 catheter-days, the rate is:

(8 ÷ 2,000) × 1,000 = 4 infections per 1,000 catheter-days.

This calculation assumes that catheter-days are measured consistently and that all relevant infections are identified.

Indicators may be used for internal monitoring, benchmarking, public reporting or regulatory compliance. Poorly designed indicators can encourage gaming, excessive documentation or treatment of the measure rather than the patient. A balanced indicator set should therefore combine structure, process, outcome, patient-reported and balancing measures.

3. Continuous quality improvement

Continuous quality improvement (CQI) views quality as an ongoing organisational responsibility rather than a one-time inspection. It examines systems and processes, uses data, involves those who perform the work and tests changes on a small scale before wider implementation.

The PDSA cycle

The Plan–Do–Study–Act cycle is a practical improvement method.

1. Plan: define the problem, aim, baseline, stakeholders and proposed change.
2. Do: test the change on a limited scale and record observations.
3. Study: compare results with the prediction and examine unintended effects.
4. Act: adopt, adapt or abandon the change, then begin another cycle.

A good aim is specific and time-bounded, such as: “Increase documented medication reconciliation at discharge from 62% to 90% within four months without increasing discharge delays.”

Other useful tools include process maps, flowcharts, cause-and-effect diagrams, Pareto charts, run charts, check sheets and the five-whys method. Lean methods focus on waste and flow; Six Sigma focuses on reducing variation and defects. These approaches can complement clinical judgement but should not reduce care to mechanical production.

Variation and interpretation

Variation may be common cause, arising from the normal design of a system, or special cause, arising from an unusual event such as equipment failure or an unexpected staffing shortage. A run chart displays performance over time. A control chart adds statistically derived limits, but limits are not the same as clinical standards. A process can be stable yet consistently unsafe, or unstable while having an acceptable average. Managers must therefore combine statistical signals with clinical and contextual judgement.

4. Clinical audit

Clinical audit is a structured process in which current practice is compared with explicit criteria, changes are implemented when necessary, and performance is remeasured. The audit cycle is:

1. select a topic of importance;
2. define evidence-based standards;
3. determine the population and sample;
4. collect reliable baseline data;
5. compare performance with the standard;
6. identify causes of gaps;
7. implement an improvement plan;
8. re-audit and sustain gains.

For example, an audit may examine whether patients with suspected sepsis receive timely assessment, cultures and appropriate treatment according to local policy. Audit is not merely data collection. It must lead to action and reassessment. Confidentiality, appropriate approval, secure data handling and fair interpretation are essential.

Audit differs from research. Research seeks generalisable new knowledge, whereas audit evaluates whether existing standards are being met locally. Some projects contain elements of both and should be reviewed under applicable institutional and legal requirements.

5. Accreditation and standards

Accreditation is an external assessment against published standards by an authorised or recognised body. Standards may address governance, patient rights, medication management, infection prevention, facility safety, documentation, human resources and continuous improvement.

Accreditation can strengthen accountability, standardisation and public confidence. However, it can become a “survey preparation” exercise in which documentation is improved temporarily without changing daily practice. Sustainable accreditation readiness means that safe processes are embedded, staff understand why standards matter and leaders monitor actual performance rather than relying solely on policies.

A practical standards-management system should include:

• a standards register identifying responsible owners;
• evidence of implementation, not merely policy documents;
• risk-based gap assessments;
• staff education and competency checks;
• internal tracers or walk-rounds;
• corrective-action plans with deadlines;
• executive review and follow-up.

Managers must distinguish mandatory legal requirements, professional standards, accreditation requirements and locally chosen targets. Compliance does not necessarily equal excellence, and accreditation does not eliminate clinical risk.

6. Incident reporting and patient-safety culture

An incident is an event or circumstance that could have, or did, result in unnecessary harm. A near miss reaches the patient or process boundary without causing harm, while a no-harm incident reaches the patient but causes no injury. A sentinel or serious reportable event is a particularly severe event requiring urgent investigation under applicable policy.

A mature reporting system is:

• easy to use and available at the point of care;
• based on clear definitions;
• confidential but not secretive;
• focused on learning and system improvement;
• supported by timely feedback;
• protected from retaliation for good-faith reporting;
• linked to escalation when immediate danger exists.

A just culture distinguishes human error, at-risk behaviour and reckless behaviour. Human error calls for redesign, coaching and support; at-risk behaviour calls for removing incentives and increasing awareness; reckless behaviour may require proportionate accountability. A just culture is not a blame-free culture: it combines fairness with professional responsibility.

After an incident, the immediate priorities are to care for the patient, make the environment safe, preserve relevant information, notify appropriate leaders, communicate honestly and meet disclosure obligations. Records should be factual, timely and objective. Patients and families generally deserve an open explanation, an apology where appropriate and information about corrective action, subject to law and policy.

7. Root-cause analysis

Root-cause analysis (RCA) is a structured retrospective investigation of why an incident occurred and why existing controls failed. It should ask not only “Who made the error?” but also “What conditions made the error more likely?”

A robust RCA typically includes:

1. a clear event description and scope;
2. a timeline based on records and interviews;
3. identification of active failures and latent conditions;
4. examination of communication, staffing, training, equipment, environment, policy and workload;
5. identification of contributing factors;
6. selection of corrective actions;
7. assignment of owners and deadlines;
8. monitoring for effectiveness.

The five-whys technique can expose underlying causes, but it should not be used mechanically. Complex incidents usually have multiple interacting causes. Corrective actions should follow the hierarchy of effectiveness: stronger actions such as forcing functions, standardisation, automation and physical redesign are generally more reliable than warnings, education and reminders alone.

A weak action might be “remind staff to be careful.” A stronger action might be barcode verification that prevents administration when patient, medicine or dose does not match, supported by clinical override governance.

8. Failure Mode and Effects Analysis

Failure Mode and Effects Analysis (FMEA) is a prospective method. A team maps a process, identifies possible failure modes, estimates their effects and causes, and prioritises preventive controls before harm occurs.

Traditional FMEA may assign ratings for severity (S), occurrence (O) and detectability (D), each commonly scored on a locally defined scale. A risk-priority number is sometimes calculated as:

The ratings are ordinal, not precise measurements. Therefore, an RPN should not be treated as a true probability or used as the sole basis for decisions. Some organisations prioritise high-severity failures regardless of RPN, use action matrices or supplement FMEA with clinical judgement.

For each failure mode, the team should specify existing controls, proposed actions, responsible persons, target dates and a method for verifying effectiveness. FMEA is particularly useful for new services, high-risk medicines, transfusion, surgical pathways, specimen handling and electronic-system changes.

9. Infection prevention and control

Infection prevention and control (IPC) protects patients, staff and visitors. It combines standard precautions for all care with transmission-based precautions where indicated.

Core practices include:

• hand hygiene at appropriate moments;
• risk-based use of personal protective equipment;
• respiratory hygiene and cough etiquette;
• safe injection and aseptic technique;
• cleaning, disinfection and sterilisation appropriate to the item;
• safe handling of sharps and waste;
• isolation or cohorting when required;
• surveillance of healthcare-associated infections;
• antimicrobial stewardship;
• vaccination and occupational health measures;
• education, auditing and feedback.

Hand hygiene is not a substitute for environmental cleaning, aseptic technique or device-care bundles. Managers should ensure sinks or alcohol-based hand-rub stations are accessible, supplies are available, workload is realistic and compliance observations are interpreted carefully. Surveillance definitions must be consistent, and infection rates should account for exposure, such as device-days or procedure volume.

Antimicrobial stewardship aims to ensure that antimicrobials are prescribed only when indicated, with the correct agent, dose, route and duration. It reduces adverse effects and antimicrobial resistance while preserving treatment effectiveness.

10. Medication safety

Medication-related harm can occur during prescribing, transcribing, dispensing, administration, monitoring or discharge communication. High-risk medicines include insulin, anticoagulants, opioids, concentrated electrolytes and some chemotherapy agents, although risk depends on context.

Essential controls include:

• accurate medication reconciliation at admission, transfer and discharge;
• two or more patient identifiers;
• clear, complete and legible prescriptions;
• standard concentrations and approved abbreviations;
• allergy documentation and review;
• separation of look-alike and sound-alike medicines;
• secure storage and controlled access;
• independent checks for selected high-alert medicines;
• barcode or electronic verification where reliable;
• monitoring of therapeutic response and toxicity;
• patient education and teach-back;
• reporting and analysis of medication incidents and near misses.

Technology can reduce some errors but introduces new risks, including alert fatigue, copy-forward errors, incorrect defaults, downtime and poorly integrated systems. Human review remains necessary. Medication reconciliation is particularly important because transitions of care create discrepancies between what the patient was taking, what is prescribed and what is communicated.

11. Case study: reducing wrong-patient medication incidents

This is an illustrative case. A 250-bed hospital noticed six wrong-patient medication near misses in three months on a medical ward. No patient was harmed, but staff reported interruptions, similar patient names and hurried medication rounds.

Background and problem

The existing process required a verbal name check and review of a paper medication chart. Patient identity bands were inconsistently visible, and temporary agency staff were unfamiliar with the ward layout. The incident reports initially described “staff inattention.”

Analysis

An RCA team mapped the medication process and found several contributing factors: two patients with similar names in adjacent beds, interruptions during preparation, incomplete handover, inconsistent use of a second identifier and storage of charts away from the bedside. The system relied heavily on memory and vigilance.

Approach

The hospital introduced a standard two-identifier procedure, visible identity bands, a protected medication-preparation zone, bedside chart availability, escalation for missing bands and targeted orientation for temporary staff. Barcode verification was piloted after checking network reliability and downtime procedures. Staff were invited to report near misses without punitive action, except where behaviour was reckless.

The team tracked wrong-patient near misses, compliance with two identifiers, medication-round interruptions and omitted doses as balancing measures. After three months, identifier compliance improved from 71% to 96%, and wrong-patient near misses fell from six to one. The remaining event occurred during electronic-system downtime and led to strengthening the paper backup process.

Learning points

The case demonstrates that incident reporting should lead to system analysis, not premature blame. Stronger controls, staff involvement, measurement and attention to downtime are more dependable than reminders alone. A reduction in incidents should be interpreted alongside reporting behaviour: fewer reports may indicate safer care, under-reporting or both.

12. Ethics, law and information governance

Quality programmes must respect autonomy, privacy, dignity, informed consent and non-discrimination. Patient data used for audit and surveillance should be limited to what is necessary, accessed by authorised personnel and stored securely. De-identification or pseudonymisation should be used where possible.

Managers face ethical tensions when transparency about an incident may expose the organisation to legal risk. Ethical practice requires honesty, compassion, timely communication and compliance with applicable disclosure, reporting and record-retention obligations. Investigations should protect confidentiality while preserving accountability.

Equity must be considered in indicator design. An apparently poorer outcome may reflect barriers faced by disadvantaged groups rather than inferior clinical effort. Risk adjustment and stratified reporting can reveal inequities that aggregate averages conceal. Artificial intelligence used for risk prediction or quality monitoring must be validated, explainable enough for its purpose, monitored for bias and subject to human oversight. Automation should support—not replace—clinical responsibility.

13. Implementation and managerial best practice

A hospital quality programme is most effective when it is connected to strategy, budgeting and daily operations. Senior leaders should establish a small number of priorities, define ownership, provide resources and review trends rather than isolated incidents. Frontline workers should participate in design because they understand work-as-done, which may differ from written procedures.

A practical implementation sequence is:

1. identify a patient-important problem;
2. define the population, standard and baseline;
3. understand the current process and risks;
4. engage patients, families and multidisciplinary staff;
5. select a feasible intervention;
6. test it using PDSA or an equivalent method;
7. measure outcome, process and balancing indicators;
8. address unintended consequences;
9. standardise successful practice;
10. continue surveillance and re-audit.

Common barriers include improvement fatigue, inadequate staffing, poor data quality, fragmented information systems, fear of blame, competing priorities and leadership turnover. Sustainability requires visible leadership, feedback, training, psychological safety, reliable infrastructure and integration into normal work.

Challenges and limitations

No quality method is perfect. Indicators can be distorted by coding practices or incomplete denominators. Patient outcomes are influenced by illness severity and social conditions. Audit may become a compliance ritual. RCA can become retrospective hindsight bias or individual blame. FMEA depends on the quality and diversity of the team’s knowledge. Accreditation may prioritise observable documentation over less visible culture. Electronic alerts may create fatigue, and AI may reproduce historical inequities.

These limitations do not invalidate quality methods. They require triangulation: combine quantitative indicators with case review, patient experience, staff reports, observation and professional judgement. Improvement should be judged by whether care is safer and more effective, not merely whether a form is completed.

Chapter Summary

Healthcare quality management integrates assurance, control, improvement and accountability. Donabedian’s structure–process–outcome model helps managers examine resources, care activities and patient results, while recognising that outcomes require contextual interpretation. Indicators should have clear definitions, appropriate denominators and balancing measures. CQI uses iterative testing, data and frontline participation; clinical audit compares practice with explicit standards and closes the improvement cycle.

Accreditation provides external standards but cannot replace daily safety culture. Incident reporting supports learning when it is accessible, fair and linked to action. RCA investigates incidents retrospectively, whereas FMEA anticipates failures prospectively. Infection prevention depends on hand hygiene, asepsis, environmental controls, surveillance and stewardship. Medication safety requires reliable reconciliation, identification, prescribing, dispensing, administration and monitoring systems. Ethical governance requires privacy, transparency, equity, proportionate accountability and responsible use of technology. Sustainable quality is ultimately a leadership and systems responsibility shared by the whole organisation.

Key Terms

Accreditation: External assessment of an organisation against published standards.

Balancing indicator: A measure used to detect unintended consequences of an intervention.

Clinical audit: Comparison of current practice with explicit standards followed by improvement and remeasurement.

Clinical governance: The system for maintaining and improving clinical quality and accountability.

Common-cause variation: Normal variation produced by the usual design of a process.

Continuous quality improvement: Ongoing, data-informed improvement involving those who deliver care.

Donabedian model: Framework examining structure, process and outcome.

FMEA: Prospective analysis of possible failures, their effects and preventive controls.

Healthcare-associated infection: Infection acquired in connection with healthcare delivery.

Incident: An event or circumstance that caused or could have caused unnecessary harm.

Indicator: A defined measure used to signal performance or quality.

Just culture: An approach combining learning from error with fair, proportionate accountability.

Near miss: An incident that does not result in patient harm despite reaching a point where harm could occur.

Outcome indicator: A measure of the effect of care on patients.

Patient safety: Prevention and reduction of unnecessary harm associated with healthcare.

PDSA cycle: Plan–Do–Study–Act method for testing improvement changes.

Root-cause analysis: Retrospective investigation of contributing system factors in an incident.

Structure indicator: A measure of resources or organisational capacity.

Process indicator: A measure of whether a desired care activity occurred.

Medication reconciliation: Systematic comparison and communication of a patient’s medicines across transitions of care.

Review Questions

Short Answer Questions

1. Define patient safety and distinguish it from healthcare quality.
2. What are the three components of Donabedian’s model?
3. State the four stages of the PDSA cycle.
4. What is the purpose of a balancing indicator?
5. Distinguish a near miss from a no-harm incident.

Long Answer Questions

1. Explain how clinical governance supports continuous improvement in a hospital.
2. Discuss the strengths and limitations of structure, process and outcome indicators.
3. Describe the clinical audit cycle and explain why re-audit is necessary.
4. Compare root-cause analysis with FMEA, including when each should be used.
5. Explain the organisational practices required for effective medication safety.

Analytical / Application Questions

1. A hospital’s hand-hygiene compliance is 92%, but its healthcare-associated infection rate is increasing. Analyse possible explanations and propose a balanced investigation.
2. A wrong-dose incident occurs after an electronic prescribing update. Design an RCA approach that examines both human and system factors.
3. Design a quality-improvement project to reduce discharge medication discrepancies. Specify an aim, one process indicator, one outcome indicator and one balancing indicator.

Multiple Choice Questions with Answers

Question 1

Which element belongs primarily to the structure domain of Donabedian’s model?

A. Patient mortality

B. Hand-hygiene compliance

C. Nurse staffing and qualifications

D. Patient-reported symptom relief

Correct Answer: C. Nurse staffing and qualifications

Question 2

A hospital records 12 catheter-associated infections during 3,000 catheter-days. What is the rate per 1,000 catheter-days?

A. 2 per 1,000 catheter-days

B. 4 per 1,000 catheter-days

C. 12 per 1,000 catheter-days

D. 36 per 1,000 catheter-days

Correct Answer: B. 4 per 1,000 catheter-days

Question 3

What is the main purpose of a balancing indicator?

A. To replace outcome indicators

B. To detect unintended consequences of an intervention

C. To calculate accreditation fees

D. To identify only individual errors

Correct Answer: B. To detect unintended consequences of an intervention

Question 4

Which activity is the ‘Study’ stage of PDSA?

A. Defining the improvement aim

B. Implementing a change across the whole organisation

C. Comparing test results with the prediction

D. Writing a policy before collecting baseline data

Correct Answer: C. Comparing test results with the prediction

Question 5

Which statement best describes clinical audit?

A. It generates generalisable new scientific knowledge only

B. It compares practice with explicit standards and supports improvement

C. It investigates only serious adverse events

D. It replaces professional judgement with numerical targets

Correct Answer: B. It compares practice with explicit standards and supports improvement

Question 6

What is the primary difference between RCA and FMEA?

A. RCA is prospective and FMEA is retrospective

B. RCA investigates causes after an event, while FMEA anticipates failures

C. RCA measures patient satisfaction, while FMEA measures mortality

D. RCA applies only to medicines, while FMEA applies only to infection control

Correct Answer: B. RCA investigates causes after an event, while FMEA anticipates failures

Question 7

Which response is generally a stronger corrective action than staff reminders alone?

A. Posting a general warning notice

B. Asking staff to be more careful

C. Introducing a forcing function or standardised design control

D. Sending an informal email after the incident

Correct Answer: C. Introducing a forcing function or standardised design control

Question 8

Which practice most directly supports medication reconciliation?

A. Comparing the patient’s previous medicines with admission and discharge orders

B. Storing all medicines in one unlocked cupboard

C. Removing allergy documentation from the record

D. Allowing unapproved abbreviations

Correct Answer: A. Comparing the patient’s previous medicines with admission and discharge orders

Question 9

Which approach is most consistent with a just culture?

A. Punishing every human error equally

B. Ignoring reckless behaviour to encourage reporting

C. Distinguishing human error, at-risk behaviour and reckless behaviour

D. Preventing staff from reporting near misses

Correct Answer: C. Distinguishing human error, at-risk behaviour and reckless behaviour

Question 10

A traditional FMEA assigns severity 5, occurrence 3 and detectability 2. Using RPN = S × O × D, what is the RPN?

A. 10

B. 15

C. 30

D. 50

Correct Answer: C. 30