Quality and Safety in Anaesthesia

Published on 27/02/2015 by admin

Filed under Anesthesiology

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 1620 times

Quality and Safety in Anaesthesia

The agenda of quality improvement and patient safety is among the top priorities for individual anaesthetists, departments of anaesthesia and health care organizations. In recent years, since publication of the Institute of Medicine (IOM) report To Err is Human, there has been growing realization that a co-ordinated approach involving all layers of organizational management and clinical services is required to improve quality and patient safety.

Patient safety incidents remain a cause for concern in healthcare systems all over the world. According to IOM, as many as 98 000 deaths in the USA can be attributed to medical errors. In the UK, approximately 900 000 incidents and near misses are reported every year, of which around 2000 result in death. It has been estimated that additional hospital stay costs approximately £2 billion a year, and negligence claims amount to an extra £400 million a year.

This chapter addresses current concepts of quality and patient safety. In particular, the following topics will be covered.

QUALITY

Various attempts have been made to define quality in health care systems precisely. In the IOM’s 2001 report Crossing the Quality Chasm, six aims were proposed to define ‘what healthcare should be’. The six aims, i.e. safety, effectiveness, patient-centredness, timeliness, efficiency and equity, provide components of a working definition of quality in healthcare systems (Table 44.1). The six aims can be used as the attributes of a comprehensive quality care system, which can be continuously monitored and improved. These six aims are the cornerstones for designing and delivering a quality service from which both patients and clinicians are likely to benefit in terms of better patient care, less suffering and increased productivity and satisfaction.

Any attempts to improve quality within an organization and a department, focussing on these attributes of high quality care, should consider whether there is organizational readiness to embrace the quality improvement programme. This will be determined by the existing culture within the organization and/or department. Establishing a quality culture within a department and making it more positive is the key to the success of any quality improvement interventions with regard to their implementation, staff compliance and sustainability.

CULTURE OF QUALITY AND SAFETY

Understanding Generation of Errors: Systems Approach

In health care systems, there is still a prevalence of a ‘judicial’ approach to errors. After an incident, people are generally quick to make judgements which often result in blaming the individual most obviously associated with the incident. This approach is a big barrier to understanding the nature of adverse events and how they can be prevented. In this regard, the healthcare industry has much to learn from other industries, such as aviation and nuclear power, which now have significant track records of a robust safety culture.

One central message from the IOM report has been that, in general, the cause of preventable deaths in healthcare systems is not incompetent or careless people, but bad systems. In order to reach this level of understanding, it is important to understand the aetiology of an error in complex organizations such as hospitals (Box 44.1). A safety incident should be seen in an organizational framework of latent failures – the conditions which produce error and violation – and active failures. This concept is captured in James Reason’s model of an organizational accident. It should be emphasized that some active failures (such as simple mistakes, lapse, fall, slip) have only a local context and can be explained by factors related to individual performance and/or the task at hand. However, it is now understood that major incidents evolve over time and involve many factors.

BOX 44.1   TAXONOMY OF ERROR

LATENT FAILURES are factors which exist within an organization or process and which increase the risk of another error causing an incident. They are separated in time and often in place from the occurrence of the incident. Reason described these as organizational influences, unsafe supervision and preconditions for unsafe acts.

Organizational influences may include aspects such as training budgets and curricula, and organizational safety culture.

Unsafe supervision might be reflected by a trainee anaesthetizing a complex case near the limits of their competence without adequate consultant supervision.

Preconditions for unsafe acts include lack of robust checking procedures, near-identical drug ampoules and inadequate rest breaks for staff.

Active errors are unsafe acts which cause (or could cause) an incident. They may be errors of commission (doing the wrong thing) or omission (not doing the right thing). Various categories of active error are often described.

Execution failures: the knowledge and the intent are appropriate, but for a variety of reasons the correct actions do not ensue. These are often referred to as slips (observable actions, related to attention) and lapses (internal events, related to memory failures). They may be failures of Recognition, Attention, Memory and Selection. These are exemplified by drugs errors in anaesthesia: every anaesthetist will be able to recall events where each of these failures has happened in their own practice.

Mistakes: the action proceeds as intended, but the wrong course has been chosen. These may be:

Rule-based: prior knowledge, intuition or a protocol is available for this situation (e.g. failed intubation) but is wrongly applied. This may be by omission, too late, or by applying the wrong rule.

Knowledge-based: an unfamiliar or novel situation requires the ‘calculation’ of a solution based on the (usually incomplete) evidence. These require thought and are particularly prone to confirmation bias – evidence which supports the current model is sought, and contrary evidence is ignored.

Violations: these are deliberate choices to deviate from agreed practice (formal standing operating procedures (SOPs), or informal custom and practice). Again, these can be subdivided:

Routine violations: corners are routinely cut. This may be at an individual or departmental level. ‘Normalization of deviance’ may occur, where unacceptable practice becomes accepted practice over time. The risk is that practice becomes further and further away from good practice gradually, such that it is not noticed or dealt with until too late.

Self-serving violations: breaking the rules for personal gratification.

Situational violations: breaking the rules because (correctly) the situation demands it. There will always be circumstances when rules and procedures do not fit. A safety-conscious organization seeks to learn from these incidents.

In the generation of an incident, organizational factors are the beginning of the sequence. These create latent failures which result from the negative consequences of management decisions, and organizational strategy and/or planning. The latent failures then permeate through departmental pathways to the workplace (e.g. the operating theatre complex). Here, they create conditions which allow violations and commission of errors. The errors generated in the workplace environment may be prevented by a front-end clinician (near-miss). However, a simple active failure on the part of the clinician at this stage allows the error to produce damaging outcomes. Figure 44.1 illustrates an example: how a medication error can result from a combination of latent failures, conditions contributing to error and violations, and active failure.

In view of this knowledge, in order to facilitate quality and safety culture, it is important that, first, the healthcare organization must accept that in the vast majority of errors/accidents, ‘system’ failure has a major role to play. Second, the organization needs to be ‘open’ about it, and this openness and transparency must reflect in all their policies and procedures. Third, the organization’s response to these accidents must be ‘just’, and non-punitive for the individual involved. Finally, there must be mechanisms and forums so that ‘learning’ can occur at all levels within the organization, and so that systems are continuously improved.

Focus on Safety Behaviour and Non-Technical Skills

In clinical practice, some personal attributes of health care workers naturally render them more safe than others. These attributes are what one might see in a most highly respected member of a department.

There is now growing realization that these attributes along with non-technical skills are essential among staff to raise quality and safety in an organization. The General Medical Council in the UK makes it clear that doctors are expected to have these personal attributes. Detailed analyses of some of the high-profile incidents in anaesthesia have drawn attention towards non-technical skills. These non-technical skills (NTS) are defined as ‘The cognitive, social and personal resource skills that complement technical skills and contribute to safe and efficient task performance. They are not new or mysterious skills but are essentially what the best practitioners do in order to achieve consistently high performance and what the rest of us do on a good day’.

The underlying premise is that:

In view of these facts, and the recent knowledge about human behaviour under stressful conditions, it is important that health care organizations and workers have commitment to education and practice in non-technical skills. In particular, for the operating theatre environment, the following skills are important:

Communication and Teamwork

Some of us are better communicators and better team workers than others. However, it is now well known that, if the skills are delineated, if awareness is raised about them and tools of learning are implemented by the organization, everyone can improve these skills beyond their previous level. The best possible scenario would be when all members of the team know each other, have mutual trust, are able to discuss problems and issues openly, learn from each other and work together for a common goal. Team leaders have special responsibility for ensuring that they have complete trust in the other members of their team. Listening to each other is extremely important to gain trust. In an ideal world, team members would respect and trust each other, know each others’ strengths and weaknesses, have a low threshold for discussion and learning, and would be able to manage and allocate tasks such that the job at hand is accomplished in the most efficient and safe manner with all members of the team deriving great satisfaction from a job well done. In practice, it is difficult to accomplish this ideal. Therefore, organizations must actively explore and implement tools and training programmes which allow employees to enhance their level of communication and team working.

Pre-list briefings are an important component of this process for the whole operating team. They are tools to foster good communication, planning and learning for the whole team.

Debriefings are a complement to the briefing process (Table 44.2). The concept is that the team reviews its performance each day – did it match the plans made at the start of the day? Good practice is reinforced and areas of improvement are discussed constructively. Information from debriefings should be shared with other team members and necessary actions should be completed and fed back to the team.

Situational Awareness

This term implies broader understanding of ‘what is going on’, and how events may unfold. The term conveys more than just paying attention to the task. For example, a junior trainee anaesthetist being supervised by a middle-grade anaesthetist on emergency duty might be about to anaesthetize a patient undergoing straightforward minor emergency surgery, but also formulating a plan in the event that he/she is called to attend a cardiac arrest elsewhere in the hospital, bearing in mind that the consultant is dealing with a complicated case in another theatre and the middle-grade is very new to the environment of the hospital. The core elements of situational awareness include:

Analyses of major critical events, and experiments in simulated conditions, have shown that, under stress, individuals tend to develop ‘task fixation’, and lose an overall perspective of the situation. A typical example of task fixation would be an anaesthetist single-handedly making multiple attempts at intubating a patient’s trachea while overlooking the facts that the patient has been hypoxic for some time, that the assistant brought the difficult airway trolley into the anaesthetic room some time ago and that a very experienced airway expert and an ENT surgeon are working in close vicinity. Often in the aftermath of an adverse incident, one comes across statements such as ‘I had not realized that….’, ‘we had not anticipated that…..’, ‘it just took us all by surprise…..’.

Situational awareness refers to the ability to appraise the overall picture, to acquire relevant information quickly by scanning the whole environment and to monitor the environment continuously, in particular for any change. For example, an anaesthetist who is about to anaesthetize a patient with anticipated airway difficulty realizes that the assistant is in conversation with a colleague regarding rota cover for the evening, and the nurse is on the telephone talking to someone about a malfunctioning piece of equipment. The theatre environment, when put under stress, can expose latent conditions which produce errors. The anaesthetist would be wise to wait for a few seconds until these matters are resolved or temporarily postponed, and ensure that all relevant members of staff are not distracted from the task at hand.

Decision-Making

This is a sensitive area, because individual decisions can be deeply entangled with the individual’s professional capability. To be potentially challenged on the decisions can be intimidating for an individual, and damaging to their self-esteem. In the area of patient safety, where evidence is still emerging, education not established and procedures and protocols still not embedded, decision-making for an individual in a condition of uncertainty can be very challenging.

Decisions may have to be taken based on previous experience, knowledge, intuition and/or good prevailing sense at the time. However, the process of decision-making can be evolved and evaluated using crisis management scenarios in a simulated environment. Here, the decisions do not have clinical consequences and the evaluation can be non-threatening. Examples of training in such scenarios include, among many others, rapid sequence induction, failed intubation, unexpected severe hypotension, cardiac arrest, malignant hyperthermia and anaphylaxis. All anaesthesia departments should have working protocols to deal with crisis situations in order to help clinicians to make logical, systematic decisions under stressful conditions. The scenarios should be practised regularly in teams, and the experience of training should be enhanced by debriefing on team work and individual decisions. It is unlikely that all existing protocols will cover all possible eventualities. Therefore, decisions will still need to be made on individual choices and judgements. A culture of openness within the department should allow healthy discussion on decisions if different options could be considered, and lessons should be learnt to reinforce decision-making skills in a collective manner.

MEASURING SAFETY AND QUALITY

Safety Culture

So far, the measurement of the safety culture within an organization or a department has been mainly the subject of research projects. In the clinical arena, these concepts are now being considered essential for understanding the prevailing conditions and also to monitor the progress and success of various safety and quality initiatives or interventions.

The terms ‘safety culture’ and ‘safety climate’ have often been used interchangeably. However, there are differences which are important to understand.

According to one of many definitions, safety culture is ‘the product of individual and group values, attitudes, competencies and patterns of behaviour that determine the commitment to, and the style and proficiency of, an organization’s health and safety programmes’. In simple terms, safety culture refers to ‘the way we do things round here’; the ‘here’ in this context could be an individual, a group, a team, a department or an organization. One can also think in terms of culture being ‘what happens when no-one is watching’. A positive safety culture in an organization is characterized by its individual members respecting and trusting each other, perceiving safety to be important and having confidence that the safety interventions would be effective.

Safety climate refers to ‘surface features of safety culture from attitudes and perceptions of individuals at a given point in time’, or ‘measurable components of safety culture’.

The two terms, safety culture and safety climate, can be differentiated by comparing the ‘culture’ to an individual’s personality, and ‘climate’ to his/her mood.

The commonest method of measuring safety culture/climate in healthcare involves using quantitative questionnaires. These questionnaires have different dimensions, lengths and reliability on psychometric testing. The most commonly used tools are:

In addition to these quantitative questionnaires, many researchers use qualitative methodology including observations, semi-structured interviews and focus groups to make in-depth assessments. According to the concept of varying models of cultural maturity in healthcare settings, safety culture can be described in terms of five different levels of maturity:

Some safety measurement tools such as the Manchester Patient Safety Framework and the Patient Safety Culture Improvement Tool incorporate the concept of culture maturity, and are therefore useful in assessing changes in culture and its maturity over a period of time.

Measuring Quality

A comprehensive measurement of quality would assess, and somehow measure, all the individual components of the quality matrix, i.e. safety, clinical effectiveness, patient experience, timeliness, efficiency and equity. In 1988, Avedis Donabedian introduced a framework for assessing quality in healthcare (Table 44.3). This framework is based on three core domains – structure, processes and outcomes.

It is important to understand that the three domains of measuring quality are interdependent. Good processes depend upon good structures, and they often lead to good outcomes. Consequently, assessment of only one domain (e.g. outcome – mortality) cannot assess quality, or serve to improve it. It becomes useful only when this assessment is combined with finding the gaps in structure (e.g. staff shortage) and/or processes (e.g. non-adherence to guidelines). At present, much is being written and discussed about what should be measured to assess the existing quality of care, and any improvements. The Donabedian framework can be used to address each component of quality as shown in Table 44.4.

Clinical Outcomes: Real or Surrogate

In anaesthesia, a lot of reliance is put on real clinical outcomes to indicate quality of care, such as incidences of epidural haematoma, nerve damage, brain damage or death. Although this may sound straightforward, there are limitations associated with real outcomes.

For these reasons, adverse events and near misses are usually taken as surrogate clinical outcomes. By monitoring the surrogates, a comprehensive picture can be obtained of the work flow and processes. Overall, more than 100 outcome measures have been described which can be used to assess the quality of anaesthesia care. It is clear that monitoring all of these will require resources beyond the capacity of many individual departments. Therefore, departments and individuals will be making their own choices about what to measure.

Quality Indicators at Individual Patient Level

There is very little guidance as to what individual departments of anaesthesia should be measuring on a routine basis in every patient as part of their quality improvement programme. A framework is suggested here, which includes sentinel events, adverse events, rate-based outcome indicators and work process quality indicators (Fig. 44.2). Some of these data would be available from existing hospital IT systems; others may have to be collected as part of on-going audit or service evaluation programmes. The framework can also be used by individual anaesthetists to maintain a record of their own activity and quality of care. At the departmental level this can inform, on a continuous basis, the quality of care which the department is providing to its patients, and the effect of any quality improvement programme.

Quality Improvement Tools

Requirements for Quality Improvement: The challenge for a department, in addition to collecting all the data at an individual patient level, is to analyse the data systematically, disseminate the findings as widely as possible and identify learning outcomes and areas for improvement. In order to improve quality, departments will be required to develop interventions targeted at areas for improvement, implement them, and monitor progress and the impact of implementation on outcome measures. For a successful quality improvement programme, it is important for departments to invest in creating leaders and teams. It is vital for the leaders to address and improve the prevailing quality culture within the department. Continuing education and programmes to raise awareness are extremely important. Embedding a culture of incident reporting and learning is vital. Morbidity and mortality meetings and other network opportunities at which all quality issues can be aired and discussed without fear of a punitive outcome are absolutely essential for embedding quality consciousness.

Recent experience with implementing the WHO checklist has presented many challenges and has provided insight into what is required for successful implementation of a quality improvement intervention. Clear demonstration of success of the interventions, with the help of data and its rigorous analysis, is vital for motivating staff and increasing their belief in such interventions. Whilst all failures should be used as learning experiences, the successes must be celebrated. Barriers, facilitators and approaches in engaging staff into quality improvement are summarized in Table 44.5.

Checklists: Although they existed earlier, the modern era of the safety checklist is generally attributed to the crash of the newly designed Model 299 Boeing bomber in 1935. The cause of the crash was straightforward – the elevator lock had been left on. However, two extremely important facts emerged from the investigation. First, the pilot was an extremely experienced and competent pilot. Second, taking the elevator lock off was not an unusual requirement. The pilot would do this normally, but on this occasion, presumably due to distraction by other things, he forgot. The response of pilots was to create a process which ensured that simple things did not get missed, regardless of the situation.

The introduction of the WHO checklist as a strongly encouraged or mandated part of theatre practice has re-focussed attention on the potential benefits and risks of checklists in healthcare, and particularly perioperative care. Checklists are not new in anaesthesia. Various national bodies produced anaesthetic machine checklists in the 1980s and 1990s. This was in response to the emerging data that demonstrated a significant number of patient safety incidents which could have been avoided by proper checking of equipment.

The term checklist is used for documents which may have a variety of purposes. Conceptually, they can be thought of as:

There is a multitude of checklists in use by anaesthetists and they encompass these different purposes to varying degrees. Some are designed better than others.

Most checklists are tabulations or summaries of accepted best practice. We would hope that anaesthetists would recognize the need to undertake these actions regardless of whether a checklist exists. Their role is therefore about ensuring that these actions take place for every patient, every time. The human memory is fallible and regardless of professional status, cannot be relied upon to remember more than about seven items.

There is a substantial body of evidence to show that anaesthetists are not particularly good at completing all the necessary checks. This applies to the anaesthesia machine, drug checking and the WHO checklist. The reasons for this are multifactorial and relate to all aspects of safety and quality discussed in this chapter. Specific barriers to full engagement with checklist processes include the following.

Perceived importance. The rate of incidents related to the items on the checklists is very low and therefore the overwhelming likelihood is that there will be nothing amiss. The anaesthetist may therefore consciously, or subconsciously, choose to prioritize another activity.

Organizational culture. Perceptions of organizational prioritization of efficiency over safety may encourage an individual to save time by shortening or omitting checklists. Conversely, an organization which fails to monitor compliance with checklists, or hold individuals to account when appropriate, sends a message that it does not value them either.

Resistance to standardization. Anaesthetists are highly trained individuals with a large amount of professional autonomy. Checklists, by definition, standardize and consequently restrict practice, and may therefore be resisted. This may manifest itself in ‘academic’ arguments about the validity of evidence in favour of specific checklists.

‘Professional’ behaviour and stereotypes. As a result of the three aspects above, senior medical staff may view the checklist process as beneath them, and something to be delegated to more junior staff. This may in turn promote a culture of non-importance.

Checklist design. Undoubtedly, some checklists are badly worded and designed. There may be too many questions, questions asked at the wrong time in the process, or questions which encourage yes/no answers without true engagement. The evolution of the AAGBI Anaesthetic Machine Checklist demonstrates improvements in layout and wording to encourage full compliance.

Human fallibility. However well designed the checklist process, humans will make mistakes. Most commonly reported are automatic responses – answering yes, when the checks have not actually been done, and performing checklists by rote.

WHO Checklist: The WHO checklist is a hybrid checklist. It involves elements of planning, briefing and checking (Table 44.6). The components on the checklist match the WHO Safe Surgery Saves Lives objectives believed to be fundamental to improving outcomes following surgery (Table 44.7).

TABLE 44.7

WHO Safe Surgery Saves Lives Objectives

Objective
1. The team will operate on the correct patient at the correct site.
2. The team will use methods known to prevent harm from anaesthetic administration, while protecting the patient from pain.
3. The team will recognize and effectively prepare for life-threatening loss of airway or respiratory function.
4. The team will recognize and effectively prepare for risk of high blood loss.
5. The team will avoid inducing an allergic or adverse drug reaction known to be a significant risk to the patient.
6. The team will consistently use methods known to minimize the risk of surgical site infection.
7. The team will prevent inadvertent retention of swabs or instruments in surgical wounds.
8. The team will secure and accurately identify all surgical specimens.
9. The team will effectively communicate and exchange critical patient information for the safe conduct of the operation.
10. Hospitals and public health systems will establish routine surveillance of surgical capacity, volume and results.

There are three phases to the checklist:

Local adaptation is encouraged, both to match local circumstances and to ensure the appropriate questions are asked. For instance, cataract surgery under local anaesthesia is ill-served by questions related to airway management, but may have specific issues around preoperative biometry.

There are other surgical checklists, notably the SURPASS system, pioneered in the Netherlands. The basic principles are the same.

There is reasonable evidence that implementation of the WHO checklist is associated with improvement in outcomes following surgery. The degree of compliance with perioperative checklists appears to be associated with both intraoperative teamwork, and postoperative outcomes. In other words, doing it well is as important as doing it at all.

As with all checklists, lack of engagement, over-familiarity and ‘tick-boxing’ are significant risks from its continued use.

The WHO checklist is a per patient checklist. Many authorities argue that its safety role is enhanced by broader per list processes such as briefings and debriefings which promote wider team communication and allow problems to be identified and addressed without the pressure of having a patient present.

DESIGNING PROBLEMS OUT OF THE SYSTEM

Human error is inevitable. A key approach to minimizing error is therefore to limit the possibility for human error. In many respects, anaesthesia provides exemplars of how this can be done, although there are many areas in which progress could be made.

Learning from Incidents

A key indicator of a safe organization is its desire and ability to learn from previous incidents and near-misses. To quote Sir Liam Donaldson, a past UK Chief Medical Officer: ‘To err is human, to cover up is unforgiveable, and to fail to learn is inexcusable.’

There are several key components to learning from incidents and near-misses:

Appropriate Analysis of Incidents

Large industrial companies have specific teams trained to collate, analyse and investigate incidents. Such an approach is relatively rare within healthcare. The anaesthetic profession has made significant progress in this area with prospective studies of rare events (such as the UK National Audit Projects investigating complications of neuraxial blocks, airway management and unintentional awareness); analysis of legal claims (US Closed Claims, NHLSA database in the UK); and prospective studies of incidents (AIMS study – Australia). Many countries now have some form of national reporting system (e.g. the National Reporting and Learning System), although these are of course only as good as the data entered.

Analysis of individual incidents and aggregated data is a skill. This requires training to ensure that the correct interpretation of events is made and that any recommendations will reduce the risk of recurrence without introducing new risks.

Departmental morbidity and mortality meetings provide a good opportunity for colleagues to learn from incidents, provided they are facilitated well.

Models for Implementing Quality Improvement Programmes

Different models used in quality improvement programmes involve identifying an area for improvement, carefully defining an improvement and developing tools which can be used to measure improvement. A multidisciplinary team then considers and develops the interventions aimed at making these improvements; interventions are implemented and progress is monitored. If the intervention leads to improvements, it is then integrated into everyday work flow. A clear understanding of the aims and positive workplace quality culture are required for the success of any such programme, which requires an integrated approach involving all stakeholders at organizational, departmental and individual levels.

Some of the models of quality improvement which are often used in healthcare settings are described below.

Six Sigma: The main aim of this model is to improve processes to minimize or eliminate waste. The improvement is measured by assessing process capability. Before an intervention is designed and undertaken, baseline assessment is carried out to establish how a process is performing. This is done by observing the process and counting the defects or areas of improvement. The defect rate per million is then calculated to a sigma metric using statistical tools. Potential solutions for improvement are then designed and piloted. The measurements of the sigma metric are undertaken again to quantify improvement in the process. This method is useful mainly for process improvements and the methodology can be undertaken on a regular basis while efforts to improve processes with multiple interventions are being undertaken in an organization.

Root Cause Analysis: This methodology is commonly used for improvements at organizational and departmental level. The analysis is triggered by an adverse event. A retrospective systematic analysis is undertaken to tease out the underlying factors which may have caused and/or contributed to the generation of the event. The spirit of analysis is the belief that systems, rather than individual incompetence, are likely to be the root cause of most problems. With an overarching systems approach to understanding the problem in a non-punitive manner, a trained multidisciplinary team investigates the event to ask a series of questions: what happened, why did it happen, what factors caused it, what contributed to it and what system improvements could prevent it. The team then makes recommendations for changes in system improvements.

SUMMARY

The attributes of quality in healthcare include safety, effectiveness, patient-centredness, timeliness, efficiency and equity. In order to achieve delivery of a high quality of care, it is essential that the organizations, departments and individual members of staff develop a positive quality and safety culture. Such a culture promotes openness and learning from everyday events without being punitive. The common understanding in this culture is that the generation of an error involves latent organizational failures, error-facilitating workplace conditions and human error at the front end. Because human error is inevitable, efforts must be directed towards strengthening the systems so that reliance on the front-end of care delivery is minimized. Any quality improvement programme requires systems which measure quality and safety. A number of methods and techniques are available to assess safety culture in the workplace. In addition, specialty-specific outcome measures should be monitored to obtain a comprehensive assessment of quality. These outcome measures should be context-specific and should include clinical outcomes as well as process measures. Sentinel events are important but do not allow continuous outcome monitoring because of their low occurrence rate. Therefore, rate-based indicators, complications and adverse outcomes need to be monitored to obtain an overall sense of the quality of care. Once the areas for improvement have been identified, departments and organizations should hold team meetings to develop and design new interventions which can narrow the gap between the existing state and the desired state of quality. For implementation of these interventions, a number of approaches, e.g. educational, research, social, may be undertaken. The tools and methodologies for implementing improvement programmes include PDSA cycles, six sigma, lean production system, root cause analysis and failure mode and effects analysis.