Use of survival analysis to determine the success of a joint replacement in Australia

South Australian SSA 28 February 2018 Meeting

The speaker at the February 2018 meeting of the SA Branch was Michelle Lorimer from the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR).  Her talk “use of survival analysis to determine the success of a joint replacement in Australia” is a piece of her current methodological research works. At the beginning of her talk, Michelle highlighted the purpose of the registry to define, improve and maintain the quality of care for recipients of joint replacement. Owned by the Australian Orthopaedic Association (AOA), permanently funded by the Australian Government, AOANJRR begun its journey in 1999 in South Australia. All public and private hospitals undertaking joint replacement in Australia are participated in the registry. A patient (procedure) is automatically opted in unless they contact the registry to opt out. The joint replacement includes hip, knee, shoulder, ankle, elbow, wrist and spinal disk. Patient details, hospitals, type, date and side of procedure, diagnosis, prosthesis catalogue, cement and surgeon code are the basic information in the registry data system, she said. Michelle also discussed about the data validation which is done by state and territory health data based on ICD-10 codes and sequential multilevel matching process.

In her talk, Michelle highlighted the data collection and management process, statistical methods and recent projects include Nested Clinical Trials, PROMs, other international registries and data linkage projects with Australian Cancer Database, MBS/PBS data and ANZDATA. The number of patients and procedures recorded by the registry between 01/09/1999 and 31/12/2016 are 862,720 individual patients, 1,237,576 procedures, 186,017 (21.6%) deaths and 33.8% of patients have had more than one procedure. The Cox proportional hazards model is one of the statistical approaches to estimate the probability of revision (outcome) along with competing risk analysis in the presence of death as a competing event.

At the final stage, Michelle showed some mechanisms of hip replacement, for example primary hip arthoroplasty, total conventional hip replacement, and total resurfacing hip replacement. In particular Michelle emphasised the femoral articulating and acetabular articulating surfaces and showed that the cumulative percent revision was about 21% for Metal/Metal compared to 4-14% for other bearing surfaces.  Factors that affect the outcome of primary total hip replacement include the type of prostheses as well as the gender and age of the patient. Women have a significantly higher rate of revision due to small femoral head sizes and even a higher rate of revision following a resurfacing procedure when a large femoral head is used, but the reason for this is not clear, Michelle said. She also mentioned that the AOA has introduced an opt out policy so that all surgeons have their data matched to patients unless they choose to remain unidentified. In addition, Surgeons can monitor the outcomes of their procedures and compare their outcomes versus all other surgeons from the yearly downloadable report.

A dinner was held right after the meeting at Jasmin Indian Restaurant, 31 Hindmarsh Square, Adelaide.

Shahid Ullah


Get the latest posts delivered to your mailbox:

Show Buttons
Hide Buttons