Data Quality In The NHS: Interview With Julie Perkin, NHS Data Quality Manager

What is it like to lead and manage the delivery of data quality services in a high-pressure environment within a UK NHS Hospital Trust?

In this interview with data quality expert and manager, Julie Perkin, we learn some of the daily challenges facing data quality practitioners in this demanding industry. 

Dylan Jones: Can you briefly describe how your career developed into that of a data quality manager within the NHS?

Julie Perkin: I was originally employed as a Medical Secretary in Paediatrics. In 2002 the Trust was implementing a new PAS software system and I was seconded over to IT to aid with its implementation. Once the system was implemented I moved with IT to the Data Quality Department where I spent the last 8 years.

Dylan Jones: Describe some of the key activities that a data quality manager in an NHS Hospital is responsible for?

Julie Perkin: Key activities you are responsible for are ensuring that all the data entered onto the PAS system is fit for purpose, entered in a timely manner and up to date. Within this you are responsible for the creation and monitoring of regular reports to ensure this is so. This was then reported back to the Operational Managers for them to action.

I was also responsible for the submission of the Commissioning Data Set. This is the information used by the Hospital Episodes Statistics, Secondary Uses System to monitor the Trust on the data targets.

One of the other issues that needs to be monitored regularly is the security around users looking at patient’s records. As the Privacy Officer I was supplied with reports on who had looked at which patients and I ensured that there was a genuine need for them to be doing so.

You need to keep upto date with the Data Definitions, ensuring that all Information Standard Notices are implemented to ensure the data submitted on the CDS is complete.

Dylan Jones: What are some of the data quality initiatives you were personally responsible for implementing?

Julie Perkin: Pre-migration to the change of our PAS system, we needed to improve our NHS Number coverage, it was approximately 75%. At the time of migration I had managed a team and our coverage was 98% and the national target was 95%. At the time I left we had further continued this work at the level was 99.5%

Another change to the system we were using was the migration from using NSTS to using the Summary Care Record. I managed the migration of all users, ensuring only relevant staff were given access and that they only had the correct level of access.

Dylan Jones: In your bio you state that one of your tasks was to "Ensure that at all times the NHS Number coverage is of the highest level currently 99.5% (National expected is 95%)” – what is this data quality metric, what does 99.5% actually mean?

Julie Perkin: This metric is the number of records within you PAS system that have a "traced and verified” status. This means that the patient record is confirmed as the patient you have entered. This is extremely important with the national initiative that the first patient identifier used is the NHS Number.

It is virtually impossible to have 100% NHS number coverage as certain patients do not have an NHS number, eg, not fixed abode patients who are not registered with a GP. There are also patients who are not shown on the Summary Care Record eg, stop noted patients, gender reassignments, adopted patients.

The Trust do have the ability for a very limited number of people to allocate NHS numbers to patients who cannot be traced. I did not personally implement this system as I felt that was too much room for duplication and felt it was better done at a National Level.

Data Quality Pro: We've reported on duplicate patient identifiers occurring in the NHS before on Data Quality Pro, what are some of the main causes for duplicate ID’s being created?

Julie Perkin: There are lots of reasons why this can happen. Sometimes patients do not divulge their information, they give incorrect information. Poor quality written documents where there is a difficulty interpreting the information. Another reason can be that when users (eg, clerks) search for patients on the National Spine Database they pick the wrong patient's and then you end up with confused records. Overseas visitors, do not have an NHS number as soon as they enter the country so this may cause issues. Some people are not on the national database eg celebrities, adopted patients and this can cause problems.

Data Quality Pro: How much effort can be expended resolving duplicate ID’s?

Julie Perkin: It can take a lot of effort, phonecalls and time to resolve these issues. Confused records have to be reported nationally to ensure they are amended correctly.

Data Quality Pro: What kind of impacts can a duplicate ID have on the care team and also the patient as a result of a duplicate ID?

Julie Perkin: There are a lot of different ways it can impact the care team but if the patient's details are checked when they attend the hospital, eg at A+E or Outpatients there should be no effect for the care team. 

If it does occur, the problems can be:

  • Lack of past medical history (especially if new case notes have been created).
  • Patients listed for the wrong procedure
  • Patients not attending their appointments and could get "lost in the system".

Dylan Jones: How was your role of data quality manager perceived within the hospital? Is the value and benefit that data quality management delivers fully appreciated across the organisation?

Julie Perkin: Following the Lord Darzi’s report "Figures You Can Trust” which highlighted what Trust’s should be doing to ensure their data was fit for purpose, a lot of extra reports were created within Data Quality and the results of these were presented to the Audit Committee regularly to ensure targets had been met.

Dylan Jones: There seem to be continuous IT and data management initiatives within the NHS. What kind of impact are these directives having on data quality and trust in information?

There are forever changing goal posts and it is imperative that these are met at all times. This means a continuous changing of priorities to ensure the correct information is recorded to enable the information department to produce accurate information submissions.

Dylan Jones: What is the toughest and most challenging aspect of managing data quality at a senior level within the NHS?

Julie Perkin: It is the lack of understanding by the clerical staff inputting the data on the need for timely and accurate data to enable the patient to had a smooth and quality patient journey. 

Unfortunately due to staffing constraints clerical staff, seem to be always rushing to ensure the patients are seen and this can lead to basic errors which cause the Data Quality staff a constant stream of work to put these right.

Dylan Jones: You were responsible for Managing the Trust DQ yearly Audit Programme, can you walk us through what that entailed?

Julie Perkin: The Trust DQ Audit programme was based around auditing the reports that were submitted in the monthly Board report.

The DQ Auditor would audit the data that had been input into the Board report to ensure that procedures were being followed and data was correct.

The Audit programme does change in line with what information is required to be submitted to the Board Report.

Dylan Jones: Are there data quality strategies and frameworks that are standard within the NHS or must each data quality manager devise their own approach?

Julie Perkin: I would say each manager devises their own approach in ensuring how the data is captured and prioritise the workload to meet deadlines. 

The most important data submission for my Data Quality Department was the Commissioning Data Sets to the Secondary Uses Service (SUS) and as you can see from the link below there is access for the DQ Managers to view monthly and quarterly reports to monitor progress. 

To ensure any actions were reported back to operational managers we held a montly DQ Steering group and produced monthly Key Performance Indicator Reports (KPI’s).

Definition of terms used in this interview:      

  • PAS - Patient Administration System
  • CDS - Commissioning Data Sets
  • NSTS - National Strategic Tracing System
  • SUS - Secondary Uses Services

Below are some links to relevant web pages on topics in this interview:

Secondary Uses Services (SUS) - Data Quality Dashboards FAQ - (pdf format)

SUS - Data Quality and Operational Support

The SUS Programme (Connecting for Health website)