2017/18 Research Chairman’s report - Professor J Mark Wilkinson
NJR Research Committee
The research committee is responsible for delivering the research agenda of the National Joint Registry. Our aim is to enhance the understanding of the science of arthroplasty, improve clinical practice and benefit public health. This year we are delighted to welcome three new members to the research committee team, Mr Andrew Goldberg who brings expertise in foot and ankle surgery, Dr Jonathan Jeffers contributing his expertise in musculoskeletal biomechanics, and Mr Mike Reed who brings further expertise in lower limb arthroplasty.
Research is the route to improved treatments, and keeping patients at the centre of what we do is a guiding principle for all NJR research activity. In 2017/18 the research committee has overseen a number of significant projects which are nearing completion. Below are details of a small selection of projects amongst a varied programme of work underway. These projects represent key achievements for the NJR in meeting its aim to maximise the value of the NJR dataset to patients and the wider healthcare community. My thanks to all members of the research committee for their efforts in the past year, whose hard work makes it possible to facilitate data access and support these high quality research outputs.
Patient Decision Aid
We are pleased to announce that plans are well established to launch a web-enabled personalised Patient Decision Aid (PDA) for patients considering hip or knee replacement. This simple tool, supported by the charity Versus Arthritis, will help patients considering joint replacement make evidence-based choices about their treatment and share decision-making with their clinicians when considering the benefits and risks of undergoing joint replacement.
This project represents a substantial initiative on the part of NJR members to meet one of our core objectives to improve accessibility of the NJR resource to patients and promote shared, informed and value-based decision-making. Decision aids fill the gap between population level data and its application to the patients’ individual circumstances, to better inform patients making choices about healthcare interventions, enhance patient participation in the process, reduce decisional conflict, and benefit healthcare economies through improved clinical outcomes and better resource utilisation.
The development and validation of this decision aid has been undertaken in collaboration with arthritis patient groups who have been involved in every stage of the PDA development process, including the focus of the project, selection of input variables, presentation of the output results, and the appearance of the web-based dashboard interface. We have presented interim versions of the models to local and national patient groups and professional forums and received very positive responses. Our goal is that this work translates into a widely-utilised web-based patient resource accessible directly from the NJR and other relevant websites.
Research ready dataset and Data Access Portal
Another important milestone for the NJR is the delivery of our ‘research-ready’ dataset and online Data Access Portal which will launch later this year. These resources will enable researchers to utilise the NJR Annual Report dataset more effectively as a research source by providing a cumulative dataset that is annually updated. This approach will foster timely remote access to the dataset for researchers with approved projects. The portal will also enhance the security of the NJR dataset and transparency of published work based on the NJR dataset. We have developed user guidance, which will available on the website when the portal is launched, to aid researchers using NJR data. This guidance also contains a useful list of online tutorials, reading materials and other resources to help researchers analysing more complex datasets for the first time.
Findings of 5-year PROMs analysis
Analysis of the 5-year PROMS data related to serial outcomes is currently taking place. An earlier analysis of the NJR Extended PROMS study was prepared using data collected for 3 years postoperatively and included in the 12th NJR Annual Report (2015). This forms part of the work of the NJR Extended PROMS Working Group, a collaboration between teams at the University of East Anglia, the University of Bristol and the University of Oxford.
Findings demonstrate the feasibility of collecting long term PROMS data, with an excellent response rate over 5 years in a representative sample. Serial measures of PROMS have been recorded in a cohort of 21,449 THR and 17,849 TKR with assessment taken preoperatively, at six months, 1 year, 3 years and 5 years. Examining the pattern of joint pain reported by individuals over time suggests that while the majority show a rapid improvement in outcome, around a fifth of patients experience their maximal improvement only after three years postoperatively. By the end of the five-year interval, around a third of patients receiving implants experiences either a relapse or show no improvement. Advancing age, high BMI, depression, multiple morbidity and social deprivation are the main factors that predict a poor outcome.
There are clear indications from this preliminary analysis that the long-term PROMS provide important information relating to implant performance that is not captured in 6-month National PROMs data. In keeping with other work in this area it is noteworthy that patient related factors (age, BMI, depression and social deprivation) appear to be the main determinants of patient outcome. This will need to be considered if the aim of the register is to equate these outcomes to surgeon, hospital or implant performance. The results justify the value in measuring long term PROMS in evaluating the patient experience following arthroplasty and further follow up of this sample to establish how these trajectories project forward and relate to the need for revision.
Use of NJR data this year
The research committee encourages external use of the NJR dataset to answer questions of clinical relevance and the NJR dataset has provided opportunities for a number of external researchers this year to develop innovative projects. Several of these independent research projects were showcased at the BOA 2017 Annual Congress. During the period 2017/18, the committee sanctioned the release of data to 20 research projects, 10 of which are externally-led projects.
The NJR dataset has been used as the key resource in 11 research papers published during this same time period, of which 8 were conducted or led by external, independent researchers.
Our internal research programme is currently addressing in depth several topics that relate to NJR priorities. In a large scale collaborative national linkage study receiving wide media coverage and published in The Bone and Joint Journal, it was possible to demonstrate the reassuring finding that the risk of cardiac failure following total hip arthroplasty was not increased in patients for who metal-on-metal implants were used, in the first seven years after surgery.
Other topics of focus include a study published in The Journal of Bone and Joint Surgery looking at the implications of introducing new technologies, often used widely without adequate supporting data, a practice that can lead to widespread catastrophic failure such as occurred with metal-on-metal (MoM) hip replacements. We determined both how revision rates would have differed if, instead of receiving MoM hip replacements, patients had received existing alternatives and the subsequent cumulative re-revision rates of the patients who did receive MoM hip replacements compared with alternatives. This study highlights the consequences of widespread and poorly monitored adoption of medical technology and why adopting new technologies without adequate supporting data must not be repeated.
A retrospective observational study was published in The Journal of Arthroplasty, investigating Trabecular metal (TM)-coated acetabular components which are increasingly used in both primary and revision total hip arthroplasty (THA). However, previous studies assessing TM acetabular components have been small single-center cohorts with most lacking a control group. This study compared revision rates following primary THA between TM and non-TM-coated acetabular components. Following primary THA, TM-coated acetabular implants had a reduced risk of revision compared with non-TM implants. Although absolute differences in revision risk were small, they may be clinically significant if TM designs were implanted in more complex cases.
Please see the NJR Research Project Portfolio here on the NJR website which catalogues all sanctioned research projects, progress reports and NJR updates in relation to research using the NJR dataset together with links to all publications related to NJR data.
Research Fellowship Scheme
We are pleased to welcome Robert Middleton our 8th appointed Research Fellow who will be investigating the association between frailty and outcomes after primary knee replacement. This programme is part of our Joint Clinical Research Fellowship scheme run in partnership with the Royal College of Surgeons of England, helping us to support high quality research in orthopaedics. Prospective fellows are encouraged to apply to our rolling programme. We encourage a diverse pool of applicants, including those from the allied health professions and non-clinical backgrounds, to apply. Look out for current opportunities on the NJR website.
Tanvir Khan, outgoing Research Fellow, completed a study looking at periprosthetic fracture and joint replacement. This is the first study to demonstrate the ‘real world’ probabilities of revision for periprosthetic fracture accounting for two competing risks; death and revisions for other indications. Study findings indicate if choosing a cementless stem, for patient groups with a higher risk of periprosthetic fracture the preferential use of stems with a collared design is advocated and caution advised in using anatomic, curved or short stems.
Research is also underway by our two current Research Fellows Jonathan Evans, whose work focuses on benchmarking of hip prostheses, and Richard Craig who is investigating improving outcomes after shoulder joint replacement surgery. To find out more about the Research Fellowship programme and links to current and published work, please see the NJR website.
Submitting a research application
The research 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. Our priority is to ensure that any research question justifies access to the data and offers potential benefit to patients. Research should align to the NJR’s priority framework and be feasible, ethical, relevant and methodologically sound.
We have a commitment to uphold the standard and consistency of work that is carried out on NJR data in line with national and international legislation. Therefore, all requests must also be approved by the HQIP Data Access Review Group for final authorisation of data release. If you would like to submit a research application, the first step in this process is to submit an expression of interest. Invited applications are reviewed by the committee quarterly.
Full details of the NJR’s research application pathway, together with detailed guidance notes for applicants, can be found at www.njrcentre.org.uk.
Please see the NJR Research Project Portfolio, which catalogues all sanctioned research projects, progress reports and NJR updates in relation to research using the NJR dataset together with links to all publications related to NJR data.
A full list of publications related to NJR data is included in Appendix 4.