One of the ways to make sure that common operations, like hip and knee replacements, are being done well is to check how often problems like infections happen after them. In England the number of infections after surgery can be monitored for up to 17 specific types of operations. Data collection is a manual process with nurses checking records and following up patients and can be time consuming. So NHS hospitals are only required to collect information for just one of four types of surgery, for only 3 months every year – everything else is optional. This makes it difficult to get a good picture of how different hospitals are doing.
We want to find out whether we can use electronic health records routinely collected from patients having these operations to get “good enough” information from different hospitals. We know that the information from these records will not be quite as good as what nurses can find out on the ground, because all the details are not always routinely recorded. But if it is good enough, we could check a much larger number of operations all year round, which could help spot problems earlier, with less burden on the healthcare system. We will also look at factors relating to how operations are done, such as how a patient’s temperature changes when they are having their operation, and when do they get preventative antibiotics, to see whether changes in these can help us spot problems with infections after operations faster.
We would also like to test whether a wider scan of electronic health records can add useful information to work out which patients have highest risks of infection.