Skip to main content
Print

User Guide for Reviewing Clinical Variation

Our user guide provides a six-step approach on how to review clinical variation data and implement Action 1.28 of the National Safety and Quality Health Service (NSQHS) Standards.

Why measure clinical variation?

Substantial variation in healthcare outcomes or processes is an alarm bell that should make us stop and investigate whether appropriate care is being delivered.

Variation in itself is not necessarily bad. When it reflects differences in patients' needs, it can be an indicator of good quality healthcare. When it doesn't reflect patients' needs, it is 'unwarranted' and represents an opportunity for a health service to improve its performance.

Action 1.28 of the NSQHS Standards requires health service organisations to identify potentially unwarranted variation and regularly review and improve the appropriateness of clinical care.

This guide explains how health service organisations can implement Action 1.28 of the NSQHS Standards, which requires a health service organisation to collect data on its own clinical care processes and outcomes, and review their performance in comparison to:

  • other health service organisations, and/or
  • evidence-based guidelines or clinical care standards.

These two comparisons can also be framed as questions:

  • How does care delivered in this organisation compare with care in similar organisations?
  • How does care delivered in this organisation compare with best practice care?

Health service organisations can choose to examine clinical variation at the level of clinical teams within the organisation and/or at an organisational level. Examining both levels is recommended.

Clinical leaders and clinical teams should be involved throughout the entire process of examining variation in clinical care.

Projects from small health service organisations can be simpler and targeted on one specific area of safety and quality. For example, they may involve auditing one clinical area against one aspect of an evidence-based guideline or a clinical care standard that is regularly used by the health service.

Smaller, simpler projects are appropriate for day procedure services and small hospitals (generally under 50 beds).

Step 1: Select priority areas

Select areas of clinical care to investigate based on:

  • high volume of patients
  • high risk to patients, regardless of volume
  • high morbidity, mortality or patient dissatisfaction
  • clinical areas identified in the health service's risk register as high or potentially high risk
  • existence of an established evidence base for best practice
  • evidence that overuse or underuse of the intervention increases risk to patient health
  • advice from clinical leaders in the health service organisation
  • availability of external data or standards/guidelines for comparison with the health service organisation's practice
  • clinical areas where new evidence or technology has substantially changed the standard of care
  • interventions identified as low-value care.

Decision making about selecting the priority areas to investigate can be aligned with the organisation's risk management approach (Action 1.10 of Standard 1, Clinical Governance).

Potential clinical topics for investigation

The number and type of clinical areas selected should reflect the size and range of clinical activity, and the extent of their risks, in the health service organisation.

We have a list of potential clinical topics for investigation and sources of evidence-based recommendations.

Related resources

Step 2. Plan the project

Once you have chosen the clinical area, you will need to decide whether you will compare your data against data from other health service organisations, and/or evidence-based guideline recommendations or a clinical care standard. This decision will be influenced by the availability of these comparators.

You will also need to decide whether you will use clinical team data, departmental-level data and/or health service organisation-level data, for comparison with external data, guideline recommendations or standards.

In some cases, there may be substantial variation within the health service organisation even if the data does not show marked variation at a whole-of-organisation level compared with external data or with guidelines. For this reason, exploring the data at the level of clinical teams as well as whole-of-organisation level is very useful for ensuring appropriate care.

External data sources for comparison

There are many sources of external data on clinical care that health service organisations can use for comparison with their own data.

Reports that include data on aspects of clinical care delivery and appropriateness of care are available for many states and territories such as:

Victoria

Safer Care Victoria

New South Wales

Bureau of Health Information (BHI)

Western Australia

Safety and Quality Indicator Set (SQuIS)

The Western Australian Department of Health provides WA public hospitals access to the Safety and Quality Indicator Set (SQuIS), which shows variation and outliers. This dashboard is currently only available to staff working within WA Health. To access SQuIS, WA Health staff should be logged onto a PC on the WA Health Network, then enter /safetyandquality at the end of the URL address in their internet browser.

Clinical quality registries monitor the appropriateness and effectiveness of specific types of care.

The data collected are used to identify benchmarks and variation in clinical outcomes and the information is provided back to health professionals to help improve clinical practice.

For example, the Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) produces the Cardiac Surgery Database with details of all adult cardiac surgery undertaken in participating hospitals.

We developed the register of clinical registries, enabling health service organisations to produce a comprehensive list of all clinical quality registries to which they could contribute data.

Some specialty groups conduct clinical audits across several health service organisations.

Two examples are:

  • The Stroke Foundation – Conducts a biennial audit of acute services and rehabilitation services for stroke; produces annual reports on the use of stroke clinical guidelines and on the processes and outcomes of care for patients included in the Australian Stroke Care Registry. These reports identify the levels of care achieved on average and by the highest performing 15% of services.
  • The Victorian Department of Health – In partnership with the Australian Red Cross Lifeblood (formerly Blood Services), conducts audits of services related to blood management. The data is published in Audit Reports covering a variety of topics.

Several organisations offer benchmarking programs to members, as well as opportunities to exchange knowledge with peers, for example:

Patient Reported Outcome Measures (PROMS) reflect the patient's assessment of how clinical care and health care interventions have affected their quality of life, daily functioning, symptom severity and other dimensions of their health.

Routine reporting of PROMS is not yet embedded in the Australian healthcare system, but in some areas reports on PROMS are available. For example, the Palliative Care Outcomes Collaboration (PCOC) provides patient outcome reports every six months, including state/territory and national benchmarks, to participating clinical services.

Our Australian Atlas of Healthcare Variation series maps age- and sex-standardised variation in use of health care according to where people live.

For example, data showing the variation in age- and sex-standardised rates of both public and private hospital admissions for specific conditions or procedures, the rates of dispensing for medicines included in the Pharmaceutical Benefits Scheme, and the rates of tests and procedures subsidised by the Medical Benefits Schedule are available. Health service organisations with defined catchment areas can use these reports to examine use of health-care interventions in their area.

Health service organisations can also use Atlas data for comparison with their own data. Choosing an area with similar characteristics to its own, such as socioeconomic status and rurality, will make the comparison more useful. Atlas data is not available for individual health service organisations because the analyses are based on patient residence rather than where they are treated.

A range of clinical care standards and evidence-based guideline recommendations are available for comparison with clinical practice. Clinical leaders and managers should identify which specific aspects of a standard or guideline will be assessed in the health service organisation.

Our clinical care standards define the care people should be offered for a specific condition, regardless of where they are treated in Australia. Each clinical care standard published by the Commission contains a small number of quality statements that align with current best evidence, and a set of recommended indicators to monitor implementation.

Clinical practice guidelines often have many recommendations covering different aspects of care. These recommendations usually vary in the strength of the underlying evidence, the strength of the recommendation itself, the likely impact on patient outcomes and the extent to which adherence to the recommendation can be easily measured. These factors should be considered when health professionals and health service organisations are choosing which aspects of guideline adherence they will monitor.

Recommendations by the Royal Australasian College of Physicians’ (RACP) Evolve program can also be used for assessing variation in care:

Step 3. Measure and review

This step requires health service organisations to have processes that:

  • measure clinical care delivered and outcomes achieved in the health service organisation
  • compare the care delivered or outcomes achieved within the health service organisation to those of other organisations, using external reports, clinical quality registries and audits, and/or
  • assess the extent to which care delivery and outcomes align with evidence-based guideline recommendations and clinical care standards
  • assess the clinical importance of any noted variation and take action if required.

Determine the measures or indicators

Use a clear, consistent definition of the measures or indicators to be assessed to allow meaningful comparison with other health service organisations’ data and with guidelines or standards.

In many instances there are existing high-quality indicators that can be used to consistently measure data over periods of time, and that allow comparison with results from similar health service organisations.

For example, our clinical care standards include indicators that can be used.

Define the target population

Other key requirements include a definition of the target population – the group of patients who should be offered the specific type of care outlined in our clinical care standards or guideline recommendations – and identification of the benchmarks that should be achieved within the health service organisation.

Audit the clinical records

Clinical records can then be audited using the systems established for this purpose. Action 1.16 of the NSQHS Standards require health service organisations have healthcare record systems that support systematic audit of clinical information. If electronic systems are used to extract information, it should be possible to determine whether care was provided to all eligible patients within a certain time period.

However, if manual collection of information from medical records is required, it may only be feasible to collect information from a sample of patients. Once the information has been collected it can be summarised, and the extent of any identified gaps between best evidence and current practice can be assessed. See Using data for quality improvement.

Using data for quality improvement

There are a number of excellent sources of information on how to collect and assess data in support of a quality improvement initiative, including the below.

A hypothetical example of collecting data on clinical care can be seen in the case study Increasing prescription rates for preventative medication after cardiac events.
 

Step 4. Explore reasons

Auditing the relevant aspects of patient records is an effective way to gain insights into how and why clinical care is deviating from best practice.

Ensuring relevant staff discuss the findings as a group will draw out different perspectives on processes that affect clinical care and can help foster a team approach to improving practice.

Reasons for clinical variation

There are many reasons for variation in rates of care processes and in outcomes of care. Variation can be due to differences in the needs of patients, in which case it is warranted and desirable. Other reasons for clinical variation include:

  • clinical care not changing in line with updated evidence
  • differences in health professionals knowledge of the latest evidence or skills related to new diagnostic or interventional procedures
  • clinical uncertainty about an intervention’s place in therapy, and the need for better data on its benefits and harms
  • inequity of access to care
  • effects of financial incentives or disincentives
  • inadequate system supports for appropriate care
  • inadequate information sharing and discussion with consumers
  • barriers experienced by Aboriginal and Torres Strait Islander peoples in accessing appropriate, culturally safe care.

It is important to give patients clear information about options, and to provide adequate opportunity to discuss this information. We have a range of resources to assist with shared decision making so patients can make properly informed choices.

For some interventions there is a lack of data on benefits and risks; in this situation it is important to discuss what is known and what is unknown, and what the other options are.

Unwarranted variation

Relatively low or high rates of use of an intervention may arise for reasons other than decisions made by individual health professionals. For example, low use of some interventions may signal a problem with access to clinical care for people who need certain tests, treatments or procedures. In this instance the health service organisation may need to examine whether services, and the workforce and resources to deliver them, are appropriately allocated given the needs of people within their catchment area.

However, unwarranted variation in use may also indicate ongoing use of an outdated method of treatment and signal the need for clinical training in new procedures or treatments.

Unwarranted high rates of interventions may also occur because of an oversupply of workforce or technology leading to lowering of clinical thresholds for undertaking these interventions.

Occasionally the investigation will highlight a pattern of problems with individual or team decision-making or skills. In this situation: consider what extra support is needed to deliver the best possible care; determine whether an external review is needed with the relevant manager; and follow the performance review processes in accordance with Action 1.22 of the NSQHS Standards (performance management).

Also implement the incident management systems (Action 1.11 of the NSQHS Standards) and open disclosure (Action 1.12 of the NSQHS Standards) when relevant.

Our resources provide detailed advice on auditing patient records and other methods of investigating reasons for clinical variation.
 

Step 5. Act to improve

Actions to improve appropriateness of care that are prompted by your variation findings should be incorporated into the organisation’s overall approach to safety and quality improvement.

Continuous quality improvement occurs through a cyclical approach:

  1. specify the desired goal for improvement
  2. explore the reasons for current practice
  3. identify the barriers or enablers for any desired change in practice
  4. make changes to health care processes
  5. monitor progress and make further changes as necessary.

Actions to improve appropriateness of care will need to be embedded in the health service organisation for improvements to care to be sustained. Our using data for quality improvement resources offer detailed advice on data-driven quality improvement methods.

‘Case for improvement’ documents

‘Case for improvement’ documents, which are available for some of the clinical care standards, provide further useful information for quality improvement. For each quality statement in a clinical care standard, the case for improvement document asks the following questions:

  • Why is it important?
  • What is known about current practice?
  • What could be achieved with more consistent application of the aspects of care described?

When possible, examples are provided showing how specific approaches or systems for implementing best practice have demonstrated measurable change.

Do not delay investigating patient care to check data
If you have found that practice within the health service varies substantially from other health service organisations or from evidence based guideline recommendations or clinical care standards, the priority – and prime responsibility – is to ensure there is no problem with patient care.

Do not delay investigating potentially suboptimal care in order to recheck data.

Step 6. Monitor and report

The health service organisation should maintain records of reviews and actions taken as a result of examining variation and appropriateness of care. Processes should be in place to report actions and outcomes, consistent with the requirements of Action 1.09 of the NSQHS Standards (measurement and quality improvements).

Health professionals have a professional responsibility to review the care they provide and to participate in efforts to improve the quality of care.

Organisations employing health professionals need to ensure that there is clinical participation in internal reviews of clinical care performance, and should encourage and provide practical support for clinical participation in relevant external clinical quality registries and audits.

Health service organisations should keep records of clinician participation in reviews of clinical variation.

Health professionals can use these records to demonstrate compliance with continuing professional development requirements; for example, the continuing professional requirements for medical practitioners, which aligns with Action 1.22 of the NSQHS Standards.

Requirements for continuing professional development

Stronger continuing professional development is one of the core features of the Medical Board of Australia’s Professional Performance Framework for medical practitioners which came into effect in 2023. As well as participating in educational activities to build knowledge and skills, medical practitioners are required to review their performance and measure the outcomes of care they provide.

The expectation that health service organisations review variation and appropriateness of clinical care is consistent with the expectation of the registering body that individual medical practitioners should regularly review their performance and outcomes.

Potential clinical topics for investigation and sources of evidence-based recommendations

The list below highlights a range of potential clinical topics for investigation, and related evidence-based recommendations on which to base a review of clinical variation.

Evidence of implementing Action 1.28

  • Policy documents that identify the external clinical quality systems that the health service organisation contributes to and encourages its health professionals to take part in
  • Reports on data analyses that are used to identify variation in clinical practice and areas of risk associated with variation in clinical practice
  • Reports that compare clinical practice and outcomes with those of similar services or peer organisations
  • Reports on comparative data analysis from meetings involving health professionals that identify potential reasons for any variation, further investigations that may be needed and potential areas of risk associated with variation in clinical practice
  • Comparative data analysis on clinical variation and the outcomes associated with care using external sources such as our Australian Atlas of Healthcare Variation, or data provided by, or shared with, external organisations such as clinical quality registries, peer organisations, other organisations that offer services to compare data and improve care, and state and territory health departments
  • Records of meetings where reports on clinical variation or appropriateness of practice were discussed, and health professionals assessed interventions and managed changes in practice
  • Results from auditing clinical practice against the recommended best-practice guidelines, pathways or clinical care standards, and reports on findings that are provided to all relevant health professionals, managers and committees
  • Records of clinical participation in morbidity and mortality reviews, external audits of clinical care, and external clinical registries
  • Risk management system reports that include actions to manage identified risks associated with unwarranted variation
  • Quality improvement system reports that include actions to deal with identified issues
  • Examples of improvement activities that have been implemented and evaluated to reduce unwarranted variation.

About the NSQHS Standards

Our National Safety and Quality Health Service Standards were developed in collaboration with Australian Government, states and territories, the private sector, clinical experts, patients and carers.

The primary aims of our standards are to protect the public from harm and to improve the quality of health service provision. They provide a quality assurance mechanism, which tests whether relevant systems are in place to ensure that expected standards of safety and quality are met.

There are several related actions within our Clinical Governance Standard which forms part of our wider National Safety and Quality Health Service can be used to address clinical variation.

Action 1.09 states that the health service organisation ensures that timely reports on safety and quality systems and performance are provided to:

  1. a. The governing body
  2. b. The workforce
  3. c. Consumers and the local community
  4. d. Other relevant health service organisations.

Action 1.10 states the health service organisation:

  1. Identifies and documents organisational risks
  2. Uses clinical and other data collections to support risk assessments
  3. Acts to reduce risks
  4. Regularly reviews and acts to improve the effectiveness of the risk management system
  5. Reports on risks to the workforce and consumers
  6. Plans for, and manages, internal and external emergencies and disasters

Action 1.11 states the health service organisation has organisation-wide incident management and investigation systems, and:

  1. Supports the workforce to recognise and report incidents
  2. Supports patients, carers and families to communicate concerns or incidents
  3. Involves the workforce and consumers in the review of incidents
  4. Provides timely feedback on the analysis of incidents to the governing body, the workforce and consumers
  5. Uses the information from the analysis of incidents to improve safety and quality
  6. Incorporates risks identified in the analysis of incidents into the risk management system
  7. Regularly reviews and acts to improve the effectiveness of the incident management and investigation systems

Action 1.12 states that the health service organisation:

  1. Uses an open disclosure program that is consistent with the Australian Open Disclosure Framework
  2. Monitors and acts to improve the effectiveness of open disclosure processes

Action 1.16 states that the health service organisation has healthcare records systems that:

  1. Make the healthcare record available to clinicians at the point of care
  2. Support the workforce to maintain accurate and complete healthcare records
  3. Comply with security and privacy regulations
  4. Support systematic audit of clinical information e. Integrate multiple information systems, where they are used

Action 1.22 states that the health service organisation has valid and reliable performance review processes that:

  1. Require members of the workforce to regularly take part in a review of their performance
  2. Identify needs for training and development in safety and quality
  3. Incorporate information on training requirements into the organisation’s training system

Action 1.28 states that the health service organisation has systems to:

  1. Monitor variation in practice against expected health outcomes
  2. Provide feedback to clinicians on variation in practice and health outcomes
  3. Review performance against external measures
  4. Support clinicians to take part in clinical review of their practice
  5. Use information on unwarranted clinical variation to inform improvements in safety and quality systems
  6. Record the risks identified from unwarranted clinical variation in the risk management system.

Definitions

Appropriate care means that patients are receiving the right care, and the right amount of care according to their needs and preferences, at the right time. The care offered should also be based on the best available evidence.

Clinical variation is a difference in healthcare processes or outcomes, compared to peers, or to a standard, such as an evidence-based guideline recommendation. It is important to consider that clinical care can also vary within a health service organisation, for example, between clinicians, departments or sites. Examples of clinical variation include:

  • A higher rate of treating heavy menstrual bleeding with hysterectomy, and a lower rate of less invasive treatments, by one clinical team compared with others in the same hospital
  • A lower rate of patients with stroke who have a comprehensive discharge plan completed in one hospital, compared with the national average and the top 15% of best-performing hospitals
  • Not prescribing secondary prevention medicines for all eligible patients who have had an acute coronary syndrome, as in the Acute Coronary Syndromes Clinical Care Standard. See Appendix 1 for links to each clinical care standard.

Clinical variation is unwarranted when it does not reflect a difference in the patients’ clinical needs, or the patients’ preferences. If clinical variation does not reflect a difference in patients’ clinical needs or preferences, it is unwarranted and may present an opportunity for the system to improve.

Last updated: 13 March 2026