Answers questions that patients, carers and the general public may have when viewing ACPGBI's bowel cancer Clinical Outcomes Publication.
Information on the results (outcomes) of individual surgeons working in coloproctology (disorders of the rectum, anus and colon) for an operation for bowel cancer has been collected and released. The data relate to patients diagnosed with bowel cancer during the period between 1st April 2011 and 31st March 2016. In addition, other measures are reported for hospital trusts for patients diagnosed 1st April 2015 to 31st March 2016.
- What data are being published?
- Why are these data being published?
- Who is involved in collecting the data?
- Why is information on this one operation being used?
- Why is 90 days taken as the time?
- Why are only planned operations being looked at?
- Why can’t I find a particular surgeon in this information?
- What makes information submitted to the Audit produce correct results?
- Are the data correct? Are there any problems with the data? Why?
- What has been the result of the data collection?
- Why is there such variation in the outcome data between surgeons?
- What else is important for good results and recovery for patients?
- What does the data not tell us as patients?
- What will happen in the future?
- Are there any possible problems in collecting this data which will affect patient care?
- Patients Used in Analysis
What data are being published?
The data focus on major bowel cancer resection operations (this is when a patient has cancer and the diseased part of the bowel is removed). The data are presented by NHS Hospital Trusts in England. The results show the total number of patients who underwent a planned (non-emergency) operation for bowel cancer and the number of these patients who died within 90 days of their operation. In addition, the same information is presented for individual surgeons who performed more than 10 operations for bowel cancer at that Trust. This year’s results also include two new Trust measures: length of stay (over 5 days) and 30-day unplanned readmissions. This is in addition to last year’s new Trust measures: rate of major resection and case ascertainment.
Why are these data being published?
This information is part of the Government initiative to help patients make a choice about their care (NHS Patient Choice website).
It is very important that the data are correct, so that patients can make choices based on good information.
Bowel surgeons support the transparent and open reporting of these results, and believe that these data should be made clear to patients.
Information of this sort may give clues as to why some patients do better than others and lead to better results in the future.
Research also shows that just publishing the data, even if not used in any other way, helps surgeons to examine their work and this leads to further improvements in outcome.
Who is involved in collecting the data?
NHS hospitals have been collecting data from bowel surgeons about the results of bowel cancer treatment for some years, but these data look at a much wider range of care. The National Bowel Cancer Audit has been working closely with the NHS Trusts, trying to match the names of surgeons with the operations they have carried out.
Why is information on this one operation being used?
Major resection of the bowel for cancer has been chosen as the operation to be studied because it is carried out regularly by most specialist bowel surgeons. However, it is far from ideal as the majority of surgeons do not do a sufficient number of this operation in one year to collect enough data to allow for accurate comparison between surgeons.
Why is 90 days taken as the time?
The National Bowel Cancer Audit measures 90-day mortality across NHS Trusts, because it is recognised that most patients who die following bowel cancer surgery, do so within 90 days of their operation.
Why are only planned operations being looked at?
When patients are rushed into hospital in an emergency, they may have less choice about their care. Also, the clinical condition that caused the emergency admission will have a strong impact on their outcome.
Why can’t I find a particular surgeon in this information?
Not all surgeons who operate on bowel cancer are included in this information. There are a number of reasons for this. Many will not have performed enough operations to be able to calculate a survival figure. Some surgeons are colorectal surgeons who specialise in other diseases of the bowel rather than colorectal cancer. Some surgeons only operate on emergency cases, which were excluded from this analysis.
What makes information submitted to the Audit produce correct results?
It is essential that data are accurate and that the Audit includes every surgeon who operates and every operation of this kind undertaken. This is called data completeness. The information put into the database also needs to be correct (valid). The National Bowel Cancer Audit is confident that as much data as can be collected has been included in the analysis, although some cases have been excluded because there was insufficient information available on these patients.
Are the data correct? Are there any problems with the data? Why?
Trust data collection was not set up to identify which specific surgeon undertook an operation. As a result, some of the data needed to match a surgeon to an operation is missing and the audit system had to exclude these cases. The fewer the number of operations recorded for a surgeon, the higher their mortality rate may appear.
NHS Trusts also collect data in many different ways and this means that while some data may be accurate, other data are not correct and require further verification. More accurate data will be included in future years.
Surgeons may also work jointly or in teams so that it can be hard to identify one surgeon as being in charge of one operation. Although this is an operation undertaken by most colorectal surgeons, very few do a sufficient number, even over the five-year period, for these data to allow reliable comparison between surgeons. Whilst a lot of work has gone into extracting the data, the results this year will not tell us the real situation and will not be completely accurate for some surgeons. However, every unusual case will be looked at and further analysis undertaken.
What has been the result of the data collection?
There is wide variation in both numbers of operations undertaken and the mortality rate (death of the patient within 90 days) between surgeons. What can be said is that on average, just over 1 in 30 patients undergoing elective surgery for large bowel cancer die within 90 days from the operation. The vast majority of surgeons have mortality figures that lie within the expected range. Whilst some surgeons may appear to have a high mortality rate, this could have arisen by chance alone because of the low number of patients included in the audit. However, there are a few surgeons whose mortality figures are higher than expected. These surgeons are given the opportunity to re-examine their outcomes and identify reasons for this result.
Why is there such variation in the outcome data between surgeons?
This can be down to one of a number of factors. It has already been stated that the accuracy of the data varies from surgeon to surgeon. Some surgeons may take high risk patients or those whose cancer is advanced, and this means that they may have higher death rates than other surgeons. Some surgeons will have had a few deaths clustered together in this time period by chance alone. Their own data on a much larger number of patients show that overall their results are acceptable. From next year we will use data collected over five years which will reduce the variation and clustering effects.
What else is important for good results and recovery for patients?
To look at the operation alone is not the whole picture. Many other specialists are involved in the care of a patient with bowel cancer, both before and after the operation and during the 90-day period after surgery. This will include such people as anaesthetists, intensive care specialists and nurses. They are all vital members of a team which works for the survival of the patient after the operation, and their good recovery.
All patients are different. Some patients will have a much greater chance of survival while others will have a greater risk of dying. This may be because they are overweight, they have another serious illness, or they are older. The data have been adjusted to take into account the greater risk in some patients, but working out the risk for an individual patient remains very difficult.
What does the data not tell us as patients?
Individual data do not necessarily tell us about how good a whole unit or department is. Patients want to know that the whole team is good and so should concentrate on Trust outcomes for this information. Also it tells us nothing about whether or not patients were treated with respect or whether or not patients could discuss their concerns with doctors and nurses.
What will happen in the future?
Surgeons will have greater involvement in the collection of data about the patients they operate on and this will give a more accurate picture of surgical outcomes. Furthermore, other aspects of a patient’s care will be analysed.
Are there any possible problems in collecting this data which will affect patient care?
It is hoped that those surgeons who now undertake the care of higher risk patients will not be concerned by publication of their results. It is also hoped that surgeons will not feel under pressure to avoid such patients and select only low risk patients. Proper risk adjustment will correct any variation in the types of patients different patients operate on.
Surgeons work closely together with other hospital staff. They are all members of a team who work hard for the survival of their patients who have undergone surgery for their bowel cancer. It is therefore something of an oversimplification to link the mortality after an operation only to the surgeons.
Patients Used in Analysis
A flowchart illustrating the filtering, inclusion and exclusion of patient records in the Audit