The first knee joint registry was initiated in Sweden in 1977 followed by other Nordic countries, New Zealand and Australia. The National Joint Registry for England and Wales (NJR) was established in 2002. The purpose of these registries was to collect data that defines, improves and maintains quality of care for patients. This information reveals important factors that help to record successful outcomes in arthroplasty surgery, which can then be shared with relevant communities.
In 2013, in a move toward greater transparency, data will be published on the performance and outcomes for some orthopaedic surgeons. Individual surgeons were asked to consent to the publication of operative and mortality data in 2013, but not all surgeons have consented. From 2014, under the mandate NHS England, all surgeon data can be freely published.
The data is provided in the form of a funnel plot. This funnel plot allows for the comparison of individual orthopaedic surgeons, units, or hospitals. It shows comparison of the variation in surgeon outcomes. The plot identifies surgeons that fall outside the confidence limits indicated on the graph.
The standardised revision ratio (SRR) is plotted against the expected number of revisions (ENR) to highlight which surgeons have a higher than expected revision rate. The SRR is the number of actual revisions divided by the ENR. An SRR of 1 is ideal; whereas an SRR of 2 means that the actual revision rate is twice as high as the expected number of revisions.
My ‘unfiltered’ funnel plot has been produced by the NJR and is displayed below:-
Registry data has had an immense and positive impact in orthopaedics. It provides:
- Quality assurance
- Standardisation and validation of data
- Identification of impact outcomes that initiate change and improvement in orthopaedic practice, which should benefit the patient community, the surgeon, and manufacturers; as well as addressing healthcare funding
- Unexpected patterns/information on prosthesis survival emerge from registry data that benefits improvement in patient outcomes
However, quality of data and inclusion/exclusion criteria, which can restrict the way in which data can be recorded and interpreted, is a critical issue:
- No data is perfect
- Data quality from source is key
- Structure of data collection – inclusion and exclusion criteria affect interpretation, as do factors that influence the way the data is cut.
- A poor record of data collection and collation has been highlighted by inaccurate recording at some paediatric cardiac centres in the UK
- A key role in the survival of implants is stratification by variables. This is not always represented in registry data
- Orthopaedic communities in Australia, New Zealand and Nordic countries own the data. Ownership is fundamentally different for the UK registry as it is controlled in the political arena and can be used as political tool. Given the lack of stratification in regard to individual surgical data, there is concern that the data may be misunderstood. These are political challenges that must be managed
- Revision of the hip replacement as an end point for failure remains relatively crude. An unrevised prosthesis does not necessarily indicate a good outcome.
In the spirit of the ‘Everyone Counts’ initiative, it is right that the public has access to this information for it assists people in making healthcare choices. However, such data should be interpreted with caution, as there are a myriad of factors that can impact upon the outcome of joint replacement surgery.
In the words of Albert Einstein:
‘Not everything that counts can be counted and not everything that can be counted counts’