DEdiCADE

How did patients like me do? A patient-centred tool (DEdiCADE) to support decision-making and long-term follow-up in elective total hip arthroplasty

Status

Analysis completed. Currently working on further publications.

Team

We had three teams working on this project: one based at the University of Oxford, one at the Geneva University Hospitals, and one at the University of York.

The project was led by Rafael Pinedo-Villanueva and Anne Lübbeke-Wolff.

The research team in Oxford included Gianluca Fabiano and Sophie Cole. In Geneva, we had the support of Didier Hanouche, Christophe Barea, Stephan Cullati, and Thomas Agoritsas. From York, Nils Gutacker also contributed to the project.

Funded by

We received funding from the Fondation Privée des Hôpitaux Universitaires de Gèneve to carry out our study.

Collaborators

What is this project about?

Many countries have registries that capture data about patients having hip replacements. Their main purpose is to know how many people are getting them, who they are (sex, age, etc.), and some also collect data about the prostheses being used, the proportions of people reporting problems, etc. Geneva has a unique registry: they have been recruiting patients and collecting data for over 20 years, including questionnaires they send to patients periodically to ask them how they are doing in terms of pain, mobility, quality of life, overall health, etc.

We undertook this project to make the conversations between patients and their surgeons more useful and informative, so that instead of surgeons sharing with patients how people who have a hip replacement do, in general, they could now tell them how previous patients like them did, in Geneva at least.

Why is this project important?

This project is important because there is much data being collected today by joint replacement registries which can be used not only for their usual purposes but also to help patients directly. It is also very important because the tool we created can provide patients and surgeons with information that they are likely to find very useful as they go through the decision to have surgery and what comes next. And it is also important because we have been able to find specific subgroups of people at greater risk of some complications, for example, which was not known before, so now this can be taken into consideration prior, during, and after surgery.

What data did we use?

We used the data collected by the Geneva Arthroplasty Registry from patients having a primary total hip replacement and participating in their registry between 1996 and 2019.

Which methods did we use?

To conduct our study, we identified subgroups of patients with similar outcomes at different time points after surgery. To identify those subgroups, we used a method called conditional inference tree analysis, which finds the variables prior to surgery (normally several, e.g. age and number of comorbidities) and their respective thresholds (e.g. 63 years of age and 2 comorbidities, respectively) that characterise each subgroup (terminal node of the tree). These are displayed like a tree with branches that split into two, potentially more than once, so that anyone could use their pre-operative characteristics to find the group of people most similar to them and see how those people did after their hip replacement.

When analysing outcomes, we separated them into four groups based on a survey we conducted of current and past patients as well as surgeons in Geneva. The outcomes we analysed were: (1) pain; (2) activity; (3) complications (infection, dislocation, fracture, revision); and (4) having the other hip replaced. We examined those outcomes at one, five, and 10 years, and in some cases even at 20 years after surgery.

What did we find?

The first thing we observed is that most people improve a lot after their surgery, but not all do. We found that the variables that characterise the subgroups change depending on the outcome and the time that has passed since surgery. We also found that for some outcomes there were few subgroups of patients performing similarly, whilst for others many subgroups could be found. The tree below shows, for example, the subgroups we found when investigating night pain. We found that 80% of people who self-rated their health as “Good” or better and had “Moderate” to “Severe” pain before they had their hip replacement reported having no pain at night 1 year after surgery. Instead, only 50% of people who thought their health was “Fair” or “Poor” before the operation, regardless of everything else, reported no night pain and more than 20% said they had either “Moderate” or “Severe” pain at night one year after the operation. A friendly booklet reporting on key findings of the project is available here.

DEdiCADE

Project outputs

The most important output for us is that this information tool is currently being used with patients who are expected to have a hip replacement at the Geneva University Hospital. We believe having this information at hand makes the discussions between patients and surgeons more meaningful, helps patients have better informed expectations, and clinicians tailor their treatment and follow-up based on the experience of previous patients like them.