Aims
Written informed consent is a crucial and mandatory process in surgical practice, exercising the patient’s autonomy in choosing to accept or decline any treatment proposed to them. Knee arthroplasty surgery is one of the most common operations performed in the United Kingdom with 357,524 primary and 50,329 unicompartmental knee replacements performed in the last three years. The aim of this study was to evaluate patient recall of the three most important surgical risks presented during consent and to compare these with the three most important surgical risks outlined by the surgeons. We postulated that a degree of agreeability of risks was an indication that the consenting process was being performed to an adequate standard in keeping with Montgomery ruling.
Methods
All patients undergoing a total or a unicompartmental knee replacement in a ring-fenced University Teaching Hospital over a six-month period were included. Patients were asked to recall the three most important risks of the operation as discussed during the consent process in clinic. The individual consultants were asked to outline the risks they highlighted as the three most clinically important. This allowed a qualitative comparison of agreeability between the patient’s perception of the biggest risks of the operation and that of their operating surgeon.
Formal statistical analysis was performed after standardisation of these risks using a generalised linear mixed effects model (GLMM) starting with a full model containing all fixed effects and discarding terms until the model fit could not be improved.
Results
A total of 206 patients were identified with 162 included in the final analysis. After variable selection, the only co-variate retained in the Poisson Generalised Linear Mixed effects model was operation type. 17% of patients could not recall any complications given to them with 15%, 20% and 48% recalling 1,2 or 3 complications. However 44% of patients had no correlation of the three most important complications compared tom their consultants and only 1 patient recalled all three of the most important ones that their surgeon highlighted.
To model the binary agreement outcome, a binomial GLMM was fitted with covariates having the same interpretation as the previous model. This showed that increasing age was statistically significantly linked to decreased chance of any agreement between patient remembered and surgeon reported risks. In addition whose previous knee replacement surgery had no correlation , a previous hip replacement was linked to increased chance of any agreement between patient remembered and surgeon reported risks.
Conclusion
This study demonstrates that a substantial proportion of patients undergoing knee arthroplasty surgery do not share a common understanding of the risks with their index surgeon. Increased age is an independent predictor of reduced risk recall in knee arthroplasty and that having undergone previous hip arthroplasty surgery improves patient recall.
