For the purpose of this blog we will be discussing clinical audit, however, in order to conduct sound clinical audit one must first understand the difference between audit and research. Audit is an evaluation of current practice, It does not attempt to alter, change or implement new methodology, simply evaluate current clinical practice. Audit V Research Audit or Research
“Research is concerned with discovering the right thing to do: Audit is ensuring it is done right” (Smith, R. 1992, Audit & Research, BMJ, 305:905-6)
So how do we set about conducting our audit?
- We must first identify the area of practice which we wish to evaluate, Is there a problem? concern? or are we just interested in determining what we currently do is effective. Using the PICO Asking Focused Questions system to write our clinical question may help to identify specific answers to our problem or concern. eg. P. in patients having day surgery endoscopy I is having a general anaesthetic C as effective as sedation O in reducing Post Operative nausea?
- Is there a standard by which we can measure this practice i.e. National standard ACHS National Standards Local guideline’s or protocols.
- The collection of data is the largest part of clinical audit and what we collect and how we interpret it is paramount. We need to determine what aspect of the practice we will evaluate, e.g. Clinical handover, this has may aspects to it, are we auditing compliance with policy, quality of handover, the amount of handovers given in a set time or one aspect of the handover such as identification checking. This brings us back to our purpose, be clear about what you wish to audit, collecting too much data is as bad as not collecting enough.
In order for our data to have a statistical significance it is important to calculate the amount of patients, documents, episodes of care we wish to include.
This article link has a useful explanation understanding sample size
My favourite tool can be found at sample size calculator
Now that we know how many patients, we need to include to make our data significant, we need to design a method of data collection. Survey is often useful when seeking qualitative data from staff and tools such as Qualtrics or Survey Monkey can provide an easy interface for staff to respond to our questions, also both these tools provide the user with a report of the findings which helps when analysing large amounts of data.
Whichever tool you use it is advisable to run a short pilot study to test if it is effective and provides the data you really need.
Is your data Quantitative or Qualitative?
Quantitative data is concerned with numerical or specific data e.g. Yes/No, Age, Gender, Blood Pressure, Blood Groups. The analysis of this type of data is performed using simple mathematical techniques and can be presented using graphs or bar charts to support your findings.
Qualitative data is usually descriptive rather than numerical e.g comments on questionnaires. This data needs to be analysed differently and carefully, as it can often be subjective and open to interpretation.It can be presented in the form of a written report or conclusion.
Regardless of the type of data collected the presentation of findings should be evaluated against the primary standard or criteria, and a conclusion drawn which identifies if the data collected meets the current standard.
Recommendations for change
In your final report, recommendations for change in practice can be suggested, This should include a strategy for implementation of the findings and a specific plan and timeframe of completion. All too often we spend weeks collecting data, formulating a conclusion and then fail to implement any change. Set yourself or the organisation SMART objectives SMART Goals for implementation of findings, and re audit in six months to test the effectiveness of your process.
Further Reading: Institue for Healthcare Improvement