Risk-benefit and costs of unicompartmental (compared to total) knee replacement for patients with multiple co-morbidities: a non-randomised study, and different novel approaches to minimise confounding.
This study will compare the risks, benefits, and costs of partial compared to total knee replacement and, at the same time, it will evaluate different study methodologies. In TOPKAT study (Total or Partial Knee Arthroplasty Trial), patients with knee damage were randomly given either partial or total knee surgery. In such studies, random allocation to different treatments provides similar patient groups that can be compared to provide evidence on the treatment effects. Even though TOPKAT will give information on both procedures, randomised studies (and particularly surgical ones) have difficulties as they: 1. are very costly and ethically challenging. 2. do not provide information on rare events and long-term effects, like complications or mortality. 3. do not include patients with a complex medical history. 4. might not be representative of real life as they are conducted within a controlled research environment. There are however increasing opportunities for efficient studies using de-identified electronic medical records from actual practice, potentially including all NHS (National Health Service) patients.
We will use routinely-collected data from hospital de-identified electronic medical records linked to two other sources: 1.the National Joint Registry, which contains information on anyone undergoing joint replacement surgery in England or Wales; and 2.the national patient reported outcomes (PROMS) database, which collects data on disease-specific outcomes and health-related quality of life before and after knee/hip replacement in the English NHS. With these data we will aim to:
1. Test the validity of the proposed methods to replicate the results from the 'gold standard' TOPKAT study.
2. Use those 'valid' methods to assess the benefits, risks, hospital costs, and cost-effectiveness of partial compared to total knee replacement amongst patients with multiple conditions (i.e. those not eligible for studies like TOPKAT).
To ensure patient and public perspective is present in the study, there will be a PPI (Patient and Public Involvement) representative evaluating the progress of the study and a PPI group that will discuss the methodology used as an alternative to the traditional randomised surgical trials.
Data sources and Fair Processing
Large datasets from the NJR alongside data from NHS Digital will be used during this study. The personal identifiers of patients who had knee replacement surgery and consented for their details to be stored in the NJR will be securely transferred to NHS Digital. NHS Digital will link the data to information it collects and extract details of all hospital admissions (HES) for the study cohort and Patient Recorded Outcome Measures (PROMS). NHS Digital will securely transfer the linked data to the University of Oxford as pseudonymised data (i.e. all identifiers will be removed and replaced with a unique patient ID). The NJR will securely transfer pseudonymised information from its records to the University of Oxford using the same unique ID so that the data can be linked by the University of Oxford and processing in such a way that researchers will not be able to identify patients.
Access patient information, and the NJR study website
For more information about how the National Joint Registry and NHS Digital use patient information, including how it is collected and how patients can opt out, follow the links below:
The lawful basis for processing: The lawful basis for processing under GDPR are Article 6(1)(e) (‘task in the public interest’) and 9(2)(j) (‘research purposes’).
Funding acknowledgement: The UTMoSt project was funded by the NIHR HTA (project ref:14/36/02)
Department of Health disclaimer The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.