What are the advantages and limitations of different quality and safety tools for health care?
The term “quality tools” is used in many different ways to refer to a method used by an individual, a team, an organization or a health system. It is most often used in a narrow sense in American texts to refer to a set of simple “continuous quality improvement” methods (CQI tools). More broadly, it includes many other safety and quality methods, frameworks, programmes or systems. Some are diagnostic methods to help decision-making, some are for intervention and change only, and some include both methods for diagnosis and intervention.
The simple tools most often reported to be used in continuous quality improvement include brainstorming, cause and effect diagrams, nominal group and Delphi technique, flow charts, histograms, control charts, Pareto diagrams, run charts, scatter diagrams, checklists, tables, and counts. Less often reported are the more complex tools: statistical process control (SPC), quality function deployment (QFD), and design of experiments study (DOE). These and other tools are also reported to be used within quality assurance or audit frameworks. The most commonly used systematic “approaches” are the continuous quality improvement plan-do-check-act (CQI PDCA) cycle, the Langley model, the team quality improvement sequence (TQIS), and different versions of the audit cycle and of patient pathway frameworks.
There is no clear evidence about which of the CQI, quality assurance or audit frameworks are most effective, but there is evidence that systematic and persistent use of a framework is necessary for results. There is evidence that statistical process control and quality costing methods are effective, but that the time and skills necessary to use them properly are greater than for many other tools. Both are disadvantaged in health care by the lack of quality data.
There are different views about whether guideline development and implementation is a quality “tool”, but it is probably the most closely studied and most common method used in health care to make quality and safety improvements. Guidelines are used in CQI projects as a method of implementing a change to practice. They are most often used as a way to convert research evidence into practical rules to follow in a local setting. Other multimethod tools include benchmarking, reengineering, and patient pathway methods. There is evidence of positive results for all, although there are mixed results, higher costs and methodological questions regarding reengineering. The EFQM (European Foundation for Quality Management) organizational assessment method, some types of clinical audit and some structured peer review methods show weak evidence of positive results.
The most often used “safety tools” in health care are incident report data collection and analysis and root cause analysis methods. Other safety tools which could be used in health care are behavioural observation, crew resource management failure mode and effect analysis, and human factors engineering design methods. No studies were found that examined whether tools were used properly, or interventions to increase their effectiveness.
The main recommendations from this synthesis of the literature are based on a few studies which give weak evidence, and on a critical assessment of the descriptive research and practitioner report literature:
- Simple continuous quality improvement (CQI) tools are useful for more effective everyday problem-solving, not just for quality improvement. All health personnel should at least know what the simple CQI tools are, and should use reports and simple costing methods to assess whether to use a tool.
- There is some evidence that guidelines, patient pathway methods, quality costing and statistical process control are effective in health care, when properly applied. More use should be made of these methods. However, the latter requires more training and guidance than other methods.
Type of evidence
Review of systematic reviews and critical assessment of descriptive research and other reports.