Mismatch between patients with colonoscopy as coded diagnosis versus patients with recalls: Case study
My PN colleague was looking at our practice recall system and how we might streamline lists and make sure that coding was correct, in order that we could easily manage mail merge recalls and put action notification in patient files. While doing this exercise, I noted that there were very few recalls in the system for colonoscopies.
We then looked at how many patients had been coded as having a colonoscopy performed versus how many had recalls.
We also looked at how many patients had family history of bowel cancer coded. From some files of people who had had colonoscopies, we noted that there was a family history noted in free text in a patient profile but not coded in a searchable way.
We checked the files of all patients who had coded colonoscopies and read the colonoscopy reports and specialist recommendations for follow-up. We coded all those with family history of bowel cancer so that we could easily search for those patients and ensure this would appear in their medical history.
I needed to carry out some backend adjustments of the recall lists via the maintenance function in the recall system used at our Leichhardt practice, especially where the doctors had free text in the ‘reason for recall’ section or there were multiple names for the same condition.
We put recalls for surveillance on all those that were indicated as needing follow-up surveillance – whether at three years or five years.
We presented the activity at the combined staff meeting to let all staff know this was happening and to engage the team.
We put the action list in all the patient files so that any health professional opening the patient file would see the action and follow-up regarding bowel cancer testing/colonoscopy. With our clinical information system, once you have put an alert in the 'action' list, this will be the first screen to open in the patient file and you cannot navigate the file until you close the box (hopefully having read, noted and actioned the alert where necessary).
We looked at the patient registration form. This had previously been amended to include family history questions for several conditions (eg diabetes, breast and bowel cancer), but these were not always being added at the new patient visit. This process was also discussed with the team to ensure that these risks were recorded and coded in a searchable way.
Invitations were sent to all patients who required screening due to family history and risk of bowel cancer to visit their GP and discuss the issue.
Recalls were sent for those who had not been added to the initial recall but who required ongoing surveillance and were due for screening.
Recalls were added for those who required future follow-up.
Although the GPs were used to adding the coding for a procedure, they were more aware of adding recalls at the time of reviewing a specialist report.
After implementing the changes to our systems, the team was more engaged in recording a coded family history for bowel cancer. Similar exercises were carried out for family history of breast cancer risk mammogram recalls.
There were other patients picked up in this exercise, where family history was not an issue but specialist- recommended recalls for follow-up had not been added.
Our senior registrar was conducting a population health project on bowel screening, and because we had a system in place where family history was coded and recalls were in place, measuring the practice’s starting point became far easier.
– Ms Karen Booth, Green Book Editorial Committee