Annual progress: Data completeness and quality

There are currently three types of organisation in England, Wales, and Northern Ireland undertaking hip, knee, ankle, shoulder and elbow joint replacement surgery:

• NHS hospitals

• Independent sector hospitals

• Independent Sector Treatment Centres (ISTC)

There are no ISTCs in Wales, Northern Ireland or the Isle of Man.

Table 1.3 below shows the proportion of reported procedures by type of provider from 2011/12 to 2016/17.

As at 31 March 2018, 2,571,503 hip, knee, ankle, shoulder, and elbow joint procedures had been submitted to the NJR.  

There were 252,251 procedure submissions in 2017/18, an increase of 9,622 from 2016/17. This represents the highest level of submissions to the NJR since the registry started to collect data in 2003. 

Operation totals
Table 1.1 below shows, by country and joint type, the number of procedures reported to the NJR for the financial years 2012/13 to 2017/18.  As for the previous five years, the number of knee replacement procedures (125,164) exceeded the number of hip replacement procedures (117,008) in 2017/18 (51.7% and 48.3% as a proportion).  The number of hip and knee procedures submitted by hospitals in Northern Ireland, having decreased significantly in 2015/16, has shown a slight increase in 2017/18.

Table 1.1:  Total joint replacement procedures entered into the NJR, 2011/12 to 2017/18, recorded by country in which the procedure took place

Operation types
Table 1.2, below, shows the number of procedures reported by type from 1 April 2012 to 31 March 2018.  Primary procedures make up 92.4% of all procedures reported during 2017/18, whilst the proportion of revisions has increased slightly to 7.64%, from 7.4% in 2016/17.  The number of ankle, elbow, and shoulder revisions as a proportion of both primaries and revisions is higher than that for hips and knees.

The linkability rate for 2017/18 was 94.0%, a slight decrease from 2016/17 (95.4%). The overall linkability for all records submitted to the NJR since 2003 is 91.1%, an increase on last year’s reported figure of 90.4%. 

The linkability rate compares the number of records submitted with the patient’s NHS number with the number of procedures recorded in the NJR.  The NHS number is required to link all primary and revision procedures relating to a single patient (NJR data is submitted for NHS number tracing and validation (this is not yet possible for Northern Ireland) and the linkability rate includes NHS numbers that have been traced subsequent to the procedure details being submitted to the NJR. 

The ability to link all operations relating to a single patient is vital in determining clinical outcomes and, without a high NHS number submission and tracing rates, the NJR’s ability to monitor clinical and implant performance is adversely affected.

Where the NHS number is missing, tracing is attempted using the NHS Demographics Batch Service.  This does, however, require the correct submission of the patient’s name, date of birth and postcode. 

Currently, there is no way of tracing records in Northern Ireland where no Health Care Number (HCN) has been submitted.  As a result, these records cannot currently be linked to other procedures for the same patient.  However, work is in progress with Northern Ireland’s Department of Health to establish such a process.  The Isle of Man uses NHS numbers and these can be traced through the existing Batch Tracing Service.

The slight decrease in the annual linkability rate may be attributable to the outcomes of the data quality audit where records have been submitted retrospectively and may not contain the patient identifiers necessary to obtain a patient’s NHS number.  This is most likely to happen for records where ‘No’ is indicated for consent.  The increase in the overall linkability rate is due to ongoing service enhancements that have enabled units to identify records with missing NHS numbers, edit them, and re-submit them. These records can then be included in the regular updates using the Batch Tracing Service.

The consent rate for 2017/18 was 92.4%, a slight increase from 2016/17 (92.2%)

The consent rate compares the number of records submitted where the patient has agreed to their personal data (see note below) being stored on the NJR database with the number of procedures recorded on the NJR. It is a requirement in England, Wales and Northern Ireland that patients give voluntary consent to have their personal data held on the NJR with their operation data. This personal information is essential to link patients’ primary and revision procedures together in order to monitor the outcomes of joint replacement surgery. Without high rates of consent, the NJR cannot achieve its goals.

Patients rarely decline consent and many units achieve consent rates of 100%. Lower consent rates in other units are generally caused by a lack of robust processes which mean that the completed consent form is not available to the person submitting the procedure details to the NJR.

Support under Section 251 of the NHS Act 2006

The NJR records consents in one of three ways: ‘Yes’, ‘No’, or ‘Not Recorded’. The support granted under Section 251 enables the NJR to collect patient details where ‘Not Recorded’ is indicated for consent. This is justified for reasons of patient safety as it enables the NJR to identify those patients who may have had a prosthesis implanted that is, subsequently, subject to either an MHRA Medical Device Alert or an MHRA Field Notice. However, this group of patients cannot be contacted for any follow up audits or research projects.

The compliance  rate from 1 April 2016 to 31 March 2017 for hip and knee submissions was 95.0% for NHS hospitals in England and Wales.  The data for the year 1 April 2017 to 31 March 2018 is not yet available. 

Prior to 1 April 2014, the compliance rate was measured by comparing the number of levies raised for implant sales against the number of procedures submitted to the NJR.  With the adoption of a subscription model in April 2014, it is no longer possible to measure the compliance rate in this way.  It is not possible, therefore, to provide a chart showing annual compliance rates over the life of the NJR.

Compliance for hip and knee procedures is compared to submissions made to the Hospital Episodes Statistics (HES) service in England and to the Patient Episode Database Wales (PEDW) service in Wales.  This comparison is based on a defined set of OPCS4 codes which can be obtained from the NJR Centre if required.

Whilst the use of HES and PEDW data enables the measurement of the rate of compliance for Trusts and Health Boards, it does not include those procedures undertaken in the independent sector and for which the ‘Operation Funding’ was indicated as ‘Independent’.

The NJR continues to explore the establishment of a data sharing agreement with the Private Healthcare Information Network (PHIN) which would enable the NJR to use Private HES (PHES) data from the independent sector to monitor compliance.

There are issues when calculating compliance for NHS Trusts where those acute trusts have commissioned joint replacement procedures directly to independent sector hospitals.  In these cases, the procedure would appear in HES but not in the NJR as the independent hospital undertaking the procedure would submit the NJR record.  Whilst the NJR can determine that a procedure will be NHS funded, it cannot be determined which trust commissioned the procedure.  Unfortunately, whilst a provider code is available within HES for outsourced procedures, this data field is rarely completed.  This may result in published trust compliance rates that are lower than expected.

About the key indicators of data quality and completeness

It is mandatory for the following providers of joint replacement surgery to collect and submit data to the NJR:

• All NHS Trusts and Foundation Trusts within NHS England

• All NHS Wales hospitals

• All hospitals in Northern Ireland performing Health and Social Care Northern Ireland (HSCNI) funded procedures

• All independent sector hospitals in England and Wales

Performance against the three key indicators of data quality (compliance, consent, and linkability) has continued to be variable throughout the year and the monitoring of, and support to, orthopaedic units has remained high.  Unit-level performance against these key indicators is provided in Part Four of the report, available as a PDF download from the right-hand side of the page.

The number of procedures submitted, with the corresponding consent rates, for each hospital are provided are provided through the NJR StatsOnline service.  Performance against all indicators is also included in an annual clinical report provided to Trust and Health Board chief executives and to independent healthcare companies.