Hip and knee prostheses allow many people suffering from joint pain to regain their freedom of movement. In a limited number of cases (1-2.2%), a devastating complication occurs; infection of the surgical site (SSI), either the wound or the prosthesis. This requires long courses of antibiotics and not seldom one or multiple additional surgeries, turning what should have been an intervention greatly improving the quality of daily life into a grave and painful burden. Additionally, the substantial resources required to fight this infection strain a healthcare system that is already under pressure. Understanding which patients are at risk and what measures can be taken to minimize the likelihood of these surgical site infections is paramount in reducing these complications. In order to understand these risk factors, data are required. The first step in gaining a better understanding is through mapping how often these infections occur and which characteristics of patient and surgery seem associated. This can be done through analysis of SSI case-characteristics compared to those of all patients undergoing the same surgery within a certain period of time, otherwise known as surveillance. In Belgium, no surveillance of SSIs is done systematically, denying surgeons, infection prevention teams, and patients insight into its incidence. It also prevents from detecting patterns in risk factors and making evidence besed recommendations for improvement, as well as from evaluating their impact post-implementation. Detecting and reporting, however, takes time, and hospitals are short-staffed. Making regular surveillance feasible, therefore, would require a system that minimizes the resource investment required, for example through an algorithm flagging possible SSI’s to an investigator while filing away patients without complications. A topic this study further evaluates. This study attempts to determine in which percentage of hip and knee prosthesis placements an infection of wound or prosthesis occurred in AZ Jan Palfijn in 2022 It also tried to compare a manual way of extraction the information required to identify infections to a semi-automated option. The files of a sample of 606 patients were analyzed for mention of, or results indicating the occurrence of infection. Additional patient and surgery characteristics were collected based on what could be identified through the literature, as possible risk factors. Risk factors do not necessarily cause a surgical wound infection, but rather are seen more often in patients that develop one, at a frequency unlikely to be explained by coincidence. In order to be able to demonstrate such association, a lot of datapoints are required. This is a challenge for low incidence occurrences such as SSI, make systematic data collection through surveillance paramount. An SSI rate of 0.8% was found, which is below the average in literature. Out of the patients included in the sample, 5 developed some form of surgical site infection (1 more superficial, the other 4 deeper), 4 requiring additional surgery. Obesity was noted as a characteristic with a trend towards association with an increased risk of infection, confirming what was found in international studies. For most risk factors however, no such confirmation could be shown. The small number of infections identified partially explaining these results, since not enough data were available for pattern recognition beyond coincidence. Manually going through patient files was time-consuming, and challenges in identifying all infections, but especially the more superficial ones, arose. Due to the way data was stored, it was not possible to develop a more automated way of identifying infections in this study. However, a specialized team in Ghent continues to work on this challenge. The data analyzed in this study could be used to help test the algorithm's accuracy. This study did manage to identify some sore spots when it came to identifying patients who developed infections. For example, there was the lack of information exchange between general practitioners and surgeons, the varying quality of reporting by hospital staff or the loss of information due to follow-up consultations happening outside the hospital. It also illustrated that surveillance would be tremendously