Annual progress: Work of NJR Committees

2013/14 RCC Network Chairman's report ~ Mr Peter Howard

The RCC network has continued in its main role of providing a broad based clinical input to the NJR, interfacing with the Regional Coordinators (RC’s) and Northgate, and with representation from HQIP. There has been further involvement of individual RCC’s in the research sub-committee, the editorial board and meetings of the data quality group. 

The review of the NJR datasets has been completed this year and are scheduled for implementation in 2014/15. Changes include the addition of:

• metastatic disease to the diagnostic fields
• patient specific instrumentation and medial/lateral unicondylar to the knee replacement proforma
• addition of specific drugs to the thromboprophylaxis section (Rivaroxaban, Apixaban, Dabigatran) and 
• the removal of the “default technique” option

This year has seen the first significant modifications to the Clinician Feedback website since its introduction five years ago, firstly as a vehicle to allow surgeons to view their data prior to publication under the consultant outcomes programme, and then later to make available five, seven and ten year survivorship data. Sequential positions on the funnel plots are now also available and to be launched soon, a personalised summary report aimed specifically at appraisal documentation. Surgeons will also soon be able to download the data of their patients who have undergone revision, and also mortality cases, and it is planned that in the near future they will be to easily download their entire data through this dedicated website, rather than through the data entry system as at present – which has a number of difficulties attached.

I am pleased that RCC network is to be maintained in its current form following the restructuring of the NJR committees, and I would like to pass on my best wishes to my successor in the post of Chairman of this committee.

2013/14 Data Quality Chairman’s report ~ Professor Paul Gregg

The group met three times in the year, with a fourth meeting comprising a full day, data quality stakeholder workshop in March 2014. This workshop was designed to review what the NJR has done to date is assessing and improving data quality and to establish the foundations of a new NJR Data Quality Strategy. The development of this strategy is one of the NJR’s objectives for 2014/15 and will seek to:

• Assess data quality requirements for the myriad use of NJR data particularly in light of new initiatives for example, publishing consultant surgeon-level information
• Measure and monitor data quality against these requirements, to ensure the NJR is maximising its impact through feedback services and new products for stakeholders

Ongoing data quality work

A first focus for the year was to continue the formalisation of a data quality process through establishment of a Data Quality Lead at each Trust, Local Health Board or independent group. These key contacts are responsible for checking, on an annual basis, the numbers of procedures performed against a list of registered procedures sent to the provider by the NJR. To date, a total of 69% of NHS providers and 74% of private hospitals have actively engaged in this process, representing an increase of around 7% and 20% on the previous year respectively. This work will now be incorporated into the overall plan for assessing NJR data quality. However, it does again highlight the importance of engagement by hospitals in contributing to improved data quality.

A separate data accuracy audit planned across 20 orthopaedic units under the direction of the Regional Clinical Coordinators (RCC) was started and each RCC is now looking at a sample set of 30 patients (15 hips, 15 knees with five of each set being revisions).

Key messages from the data quality stakeholder workshop

I was delighted to chair this workshop in my outgoing year as Vice-Chairman and pleased that the NJR collaborated widely with the orthopaedic specialist societies. The day focused on barriers to the achieving data quality as well as some potential solutions which need to address; systems and processes, methods of validation and techniques (primary and secondary) as well as engagement from those responsible for data submission and dissemination of best practice.

The strategy will be taken forward by a strategy development group, under the governance of this sub-committee and approved by NJR Steering Committee this year.

2013/14 Research Chairman’s report ~ Professor Alex Macgregor

The Research Sub-committee’s remit is to maximise access to NJR for the wider research community whilst upholding the quality of research based on NJR data. The committee takes formal responsibility for the release of data for research through an impartial and objective protocol and has oversight of the use and reporting of NJR data by research groups.

Use of NJR data this year

During the period 2013/14, the committee sanctioned the release of data to 8 external research groups. These include data to inform an in-depth NIHR-funded study of infection following joint replacement led by the University of Bristol. 

Analyses based on NJR data continue to achieve a high publication profile and include a paper in the Lancet led by the University of Bristol team documenting 90-day mortality following hip joint replacement.  The study showed a substantial secular decrease in mortality over the lifetime of the registry, and shed light on clinical management strategies that have the potential to further reduce mortality risk.

A full list of publications related to NJR data is included in Appendix 4.  The research library section of our website catalogues all NJR research data releases with their progress reports, together with links to all publications related to NJR data.

New submission dates for NJR research requests

Investigators who are considering including NJR data in external funding applications are encouraged to contact the NJR in advance of submission. The committee has set quarterly submission dates for future research applications through 2014/15.

Where it is envisaged that aspects of proposed work might duplicate or overlap with existing research on the register, before submitting a formal application for the data applicants are encouraged first to discuss their proposals both with the NJR Research Officer and with other groups carrying out similar projects.  Full details of the NJR’s research data access policy and protocols, together with detailed guidance notes for applicants, can be found at www.njrcentre.org.uk.

NJR Surgeon Outliers Chairman's report ~ Professor Paul Gregg (to 31 December 2013) ~ Acting Chairman Peter Howard (1 January 2014 onwards)

The analysis of the registry data has continued on a six-monthly basis. The sub-committee meets soon after the production of each report, and then again three months later to monitor responses. Each report has seen the identification of further potential outliers based on the funnel plot analyses. Over the last year, these have been predominantly hip surgeons with stemmed metal-on-metal hip replacements as a significant part of their practice.

Annual Clinical Report to Trusts and Local Health Boards

The Annual Clinical Report to Trusts will continue to be the mechanism that demonstrates overall performance of the individual trust compared to the rest of the registry on the same measure used in the first two reports, namely data quality, mortality and revision rates of primary procedures. Funnel plots are shown for the different sub-types of hip and knee replacement as outlined in last year’s annual report. The volumes of shoulder, elbow and ankle replacements will be included for the first time in the 2015 iterations.

In addition this year, compliance with the registry versus HES will be split between hips and knees, for both primary and revision procedures. Individual surgeons’ performance will be displayed with and without implants that have been identified as outliers, and subsequent to MHRA notification have been withdrawn from the market (Ultima TPS metal-on-metal and all ASR combinations).

Outlier methodology

The outlier methods have been reviewed by the Bristol team. The committee has discussed the use of funnel plot methodology compared with CuSum (Cumulative sum) analysis and survivorship at one, three, five and seven years after surgery to ensure that, as far as possible, any adverse change in performance is identified both correctly and expediently. It was concluded that the current method serves well, continuing to take data points beyond the 99.8% confidence limits as potential outlying data.

When new unit outliers for mortality have emerged, personal contact has been made with the clinical director ahead of data publication to enable data verification and initiation of local audit prior to data publication. Similarly new surgeon (potential) outliers are contacted at least 6 weeks ahead of notification to their unit(s).

Data quality

Concerns continue over data quality, particularly with regard to case ascertainment in the early years of the registry when compliance from some units was poor. The outgoing chair has reiterated his strong preference for independent visits to units to verify and maximise compliance and theses areas of work will be taken forward by the data quality working group.

Thanks are again expressed by the Chairman and Acting Chairman to the members of the committee for their contributions both in and out of the committee meetings, and in particular the Chairman wishes to thank Peter Howard for his work on many of the reports and analyses.

2013/14 Implant Performance Chairman’s report ~ Mr Keith Tucker

As outlined in previous reports, the Implant Performance Committee comprises members of the Scrutiny Group together with representatives from industry. The Scrutiny Group includes surgeons, representatives from the MHRA, Northgate and the University of Bristol. The NJRSC Chairman is also a member of both groups and both groups are supported by HQIP.

Examining the data

There are two main committee examinations of the NJR data to assess implant performance and identify potential outlier performance, one in Spring and one in Autumn. The Scrutiny Group meet more regularly however and they will consider any concerns about specific implants which are raised to them directly, outside of the formal data assessment process.

The Scrutiny Group continues to assess potential outlier implant performance and notifies the MHRA when there are concerns. This process is reviewed constantly and, as a result of recent reviews it is likely that more comprehensive reports will be developed for the future.  It is also planned that future reports will contain details of surgical performance, including case mix.

Notification from surgeons of poor performance

We are greatly indebted to members of the orthopaedic profession who notify us when they have concerns with the performance of specific implants. We know that the NJR database does not always pick up on devices which are ‘variations’ on a very well used device range which, overall, has excellent results. That is to say, a relatively minor modification to a well known hip or knee implant range, with the same name and manufacturer, might not be performing as well as would be wished.  If there were, for example 20,000 implantations which were excellent and 500 of a variant which were not so good, we would rely heavily on surgeons notifying us of their concerns. The NJR describes this anomaly in data analysis as ‘camouflaging’.  We continue to encourage surgeons who have noticed worrying performance to contact us directly.

Future development

As always, we are looking to improve our skills at identifying poor performance, whilst at the same time recognising good performance. We see the introduction of Unique Device Identifiers (UDI) as an opportunity for the NJR’s component database to be improved along with our capability to detect poor performance.

The NJR Implant Performance Committee also works closely with the other, international registries through the International Society of Arthroplasty Registers (ISAR) and the International Consortium of Arthroplasty Registers (ICOR).  We will review our processes in the 2014/15 year.

Changing membership

Outgoing NJR Steering Committee member Mick Borroff attended his last meeting of the Implant Performance Committee.  Mick has been closely associated with the NJR since its inception and I would like to record our indebtedness to his enormous effort in helping shape the committees and for providing a link between the NJR and the manufacturers. We welcome the newly appointed industry members, Nick Wishart and Michael Green, to the main committee.  I would also like to thank Crina Cacou, MHRA representative, for all her help and cooperation as a member of our group until 2014.

With the increasing maturity of the NJR shoulder, elbow and ankle data we are looking forward to representation from the appropriate professional societies on our committee.

2013/14 Editorial Board Chairman’s report ~ Mr Martyn Porter

The Editorial Board’s role continued in 2013/14, meeting to discuss, develop and assure the quality of data and analyses presented through the NJR Annual Report. Following the launch of the 10th Annual Report in September 2013, the group has focused its efforts on moving the reporting process forward so that it can benefit from emerging technologies. Much of our work in 2014/15 will be to develop this digital plan to ensure that the data is user-friendly and continuing to adapt to meet the needs of our readers and colleagues.

A new format for the annual report

In the development of the new NJR Reports website, members of the Editorial Board provided invaluable contributions in expanding the appeal of NJR’s annual report data. A key development has been the interactive filter options offered in the presentation of clinical activity data for all joint types. It is hoped that these intuitive, flexible options will help readers to engage with the data presented.

Promoting the 10th Annual Report findings

Other achievements for the year included the launch of the 10th Annual Report and its Public and Patient Guide (see more at [LINK]), at part of an event to celebrate the 10th anniversary year of the registry. Held at the Royal College of Surgeons England, it was a privilege to introduce its findings to surgeons, patients and orthopaedic professionals alike.

This successful event was followed by a full day of NJR-inspired sessions at the British Orthopaedic Association (BOA) Congress in Manchester, October 2013. These sessions had a keen focus on how registry data drives clinical practice. The NJR received positive feedback and we hope to offer another current and educational session as part of the event in Brighton, September 2014.

Supporting in-depth studies and the NJR Fellowship programme

The Editorial Board continued to support the in-depth study programme and were delighted to see the short-term mortality papers for hip and knee surgery achieve such acclaim in The Lancet. The studies examining infection, ethnicity and PROMs are continuing.

Additionally, the group were pleased to see significant development in Jeya Palan’s NJR Fellowship post and the progress of his studies examining the Oxford medial compartment unicompartmental knee replacement and revision total hip arthroplasty.

This section provides brief outlines of the work undertaken by the NJR Steering Committee and its sub-committees during 2013/14, the NJR’s 10th anniversary year. Although brief, these sections intend to demonstrate the ever-increasing scope of the registry’s work and the commitment required by its members to meet the needs of its orthopaedic stakeholders.

NJR Steering Committee 2013/14

The NJR Steering Committee has been acutely aware in recent years of the registry’s growing work and remit. Coupled with a maturing dataset, this has already led to a number of activities looking at the way the registry operates and how it can function better over the next decade.

The NJR’s strategic plan now includes the recommendations made from a governance workshop held in 2012, attended by members and external stakeholder alike. The 2013/14 year has seen a number of these improvements put into place including a(n):

• Revised structure, governance and operating model endorsed by the NHS England
• Appointment to the NJR’s first Medical Director post
• Extensive consultation on the appropriate format and remit of a new Medical Advisory Committee and Executive Committee to improve, respectively, engagement with the orthopaedic surgical profession and efficiency in decision-making
• Establishment of a members’ remuneration policy

New members

Since April 2013, a considerable undertaking has been the appointment of new steering committee members to replace long-standing members who concluded their terms of office, and the filling of vacant posts. Working closely with the Department of Health appointments team, the NJR management team at HQIP have successfully appointed the following members in the reporting period:

Members Mr Keith Tucker (orthopaedic surgeon member) and Mary Cowern (patient representative) have had their terms extended for a further two years until 30/04/2015 and 30/09/2015 respectively.

An additional appointment was also made in February 2014 to co-opt the British Orthopaedic Association President for their term of office on an ongoing basis. Current President Professor Tim Briggs’s position will also extend for a further year in 2015/16 in his capacity as lead of the national ‘Getting It Right First Time’ project.

At the time of writing, the announcement of two orthopaedic surgeon members and NHS management member was expected and information regarding these appointments will be available at www.njrcentre.org.uk.

As the Chairman has outlined in her introduction, the NJR must thank all of its members, past and present, for their considerable contribution to the registry – especially that of Professor Paul Gregg, NJR Vice-Chairman 2003 to 2013.

The membership of the NJR Steering Committee and membership of all sub-committees for 2013/14 can be found in online appendices. Further information can also be found in the Chairman’s introduction.