Coming up with an audit or QIP idea from scratch is hard if you aren’t in radiology already, but also hard even if you ARE in radiology. If you go to your radiology department asking for an audit idea, there’s no guarantee there will be a quick and simple audit ready and waiting for you. However, if you go with a few audit ideas, then radiologists are more likely to agree to help. Luckily, there is a huge database of audits on the RCR Audit Live website. These are freely available templates for hundreds of audits with protocols already written that you can just use in your department.
The slowest audits involve going to medical records and requesting patients’ paper notes, waiting days for them to arrive and then going through each page one by one to find data. Don’t do this. Fortunately in radiology, most data is stored electronically, so you can log in and plow through your data collection in one sitting.
Collecting data is a lot quicker with help from others. For example, you could recruit a medical student to help or team up with other people applying for radiology and each lead your own audit and help each other with data collection, thereby getting your name on each other’s audits. The other great source of help is through your hospital’s IT or PACS teams, who can help you get data or lists of patients for your audit, if you ask nicely. For example, if you wanted to audit all of the CT-guided biopsies performed in the last year, you could go through each individual biopsy list yourself, which might take a long time, or they might be able to set up a search for you and create a list of the relevant patients for you at the push of a button.
If you audit a 1 year period, that’s okay for one cycle and you’ll get a lot of data, however if you want to do another cycle, you’ll need to audit for another year, which is a huge time frame. If possible, keep audits to maximum 1 month.
This is surprisingly important, so here’s an example: if I audit the quality of CT requests from A&E, then I need to involve them as well, since they are the ones requesting scans. If we didn’t involve them, then they won’t look too kindly on us scrutinising them without their knowledge and also any changes we suggest will not be implemented. Also, they may already be auditing it, so you don’t want to duplicate work. Although involving multiple departments has its role, it does slow down your audit considerably as you will need to be liaising with them. Ideally, only audit your own department (i.e. radiology).
Background - Inpatient x-rays should be reported in a timely manner so as not to delay clinical decision-making. Reports should be produced within 1 working day according to this RCR template.
Method - Go through all of the inpatient radiograph examinations from the past week or up to 1 month (depending on how much data you want to collect) and look at the date and time that the examination was completed and when the report was authorised. The difference between the two is the reporting time and you can compare that to the 24 hour working day recommendation.
Why it's quick - This audit is good because the data are being produced electronically automatically by people reporting. You just need to collect it. You are also only auditing radiologists so you don’t need to involve other departments.
Variations - Audit of turnaround times of A+E CT/radiograph reporting, 2 week wait cancer scan reporting, you could apply this to many other situations where there is a target for the reporting time.
Background - When reporting scans, if a radiologist finds an emergency or significant unexpected finding, there needs to be a failsafe system whereby the finding is communicated back to the clinician, either by e-mail, telephone, electronic patient record, etc.
Method - Go through the reports from a fixed time period (e.g. CT scan reports over a 2 week period) and check whether any critical or significant unexpected findings were alerted to the clinician and how the alert was placed.
Why it's quick - This audit is good because it only audits radiologists and all the data is electronic. The downside is that you’ll need more involvement of the radiologist supervising as they will know more about which alerts are appropriate and which findings require an alert.
Background - A good radiology report should answer the clinical question, offer a differential diagnosis and provide advice regarding the next steps (such as further imaging and referrals).
Method - Review reports from a certain time period and situation (e.g. A&E CT, routine CT/MRI reporting, etc.) to evaluate whether they fulfil the above criteria.
Why it's quick - As for no. 2, this audit is good because it only audits radiologists and all the data is electronic. The downside is that you’ll need more involvement of the radiologist supervising as they will know more about which alerts are appropriate and which findings require an alert. You can also combine this audit with no. 2, as data are collected in the same way.
Background - Radiologists spend a lot of time at their desks and are therefore at risk of conditions associated with prolonged sitting, including repetitive strain injury, lower back pain, eye strain, etc. These can lead to musculoskeletal injuries, resulting in lost productivity and increased sickness abscence. Therefore, this is a really important audit because nobody gets taught how to adjust their seat and screen height properly. The Health and Safety Executive (HSE) has guidelines for the setup of workstations and display screens, including things like seat and screen positioning (links above).
Method - Send a survey to the radiologists in your local department asking about how often they experience pain, eye strain, etc. and then asking about their workstation setup, based on the checklist in the HSE guidelines. You can then make a poster or hold a teaching session and repeat the survey afterwards.
Why it's quick - Although I usually try to avoid surveys for audits, this is a good one because it is potentially very quick to do if you can get responses in time. Although it is not an audit about clinical performance, it is a really important issue and I think most radiologists would appreciate it.
Background - When taking biopsies using image guidance, whether that be US, fluoroscopy or CT, you want to make sure that you are actually hitting the target and that patients aren't suffering from frequent complications.
Method - Find all of the US or Fluoro or CT guided biopsies from a fixed time period and look up the histology reports to see if the sample was adequate and what the histology showed. Also read the procedure note and medical notes to see if there were any intra-procedure or post-procedure complications. Many radiologists may already be auditing their biopsies so make sure to check with them first.
Why it's quick - the data are all electronic and can be collected very quickly as it is always being produced.
Background - When inserting drains (or performing other IR procedures), it is important to audit the rate of success and complications in order to highlight any potential problems with equipment, technique, etc.
Method - Find all of the drains (or other IR procedures) from a time period and review the procedure notes and medical for any complications and also whether they needed repeat procedures. Check first whether the radiologist is already auditing this themselves or not (you can always offer to lead the audit for them - they may appreciate that).
Why it's quick - As for no. 5, the data are electronic and can be collected quickly.
Background - CTPA is the gold-standard diagnostic test for pulmonary embolism, a potentially life-threatening medical emergency. If a CTPA is performed inadequately (insufficient contrast enhancement of the pulmonary arteries), then this can lead to non-diagnostic scans and therefore a need for repeat imaging and potentially delay to the patient's treatment.
Method - The RCR template uses 210 Hounsfield Units (HU) as the CT number required in a vessel in order to distinguish a thrombus. Take the CTPAs from a fixed time period, measure the HU in a circular region of interest in the pulmonary trunk to determine the adequacy of contrast enhancement.
Why it's quick - Again, data is stored electronically and CTPAs are a commonly-performed examination, so getting enough numbers is not usually an issue.
NB: unless already familiar, you may need help from your radiologists in learning how to take the measurements. Also, this audit should involve radiographers as well (since they are taking the images), so bear that in mind as it will involve liaising with them.
The last thing to consider is whether to design your own. Use the tips at the top of the page to guide you. Speak to your local radiologists and registrars to see what the department would benefit from auditing. Good things can be those that are affecting radiology registrar training (I have completed an audit on the availability of reporting spaces for radiology registrars before, as an example) as these are also good topics to talk about at interview.