Shared learning database

Oxleas NHS Foundation Trust
Published date:
January 2014

The NICE guidelines for depression and bipolar disorder both recommend lithium as an effective treatment; patients who take lithium should have regular blood tests to monitor the amount of lithium in their blood (every 3 months), and to make sure the lithium has not caused any problems with their kidneys or thyroid (every 6 months).

Lithium monitoring is known to be sub-optimal and has been the subject of an alert by the former National Patient Safety Agency (NPSA)*. This submission describes how taking a systematic approach to overcoming barriers to best practice has had a positive effect.

* Please note that the key functions and expertise for patient safety developed by the NPSA have now transferred to NHS England.

Guidance the shared learning relates to:
Does the example relate to a general implementation of all NICE guidance?
Does the example relate to a specific implementation of a specific piece of NICE guidance?


Aims and objectives


  • To improve patient care, with respect to both treatment efficacy and patient safety, by ensuring adequate monitoring of lithium treatment in line with NICE guidance.


  • Under take a baseline audit to identify current practice.
  • Identify barriers to change.
  • Implement an intervention to improve practice; specifically development of a lithium database.
  • Re-audit to monitor progress.

Reasons for implementing your project

Oxleas NHS Foundation Trust provides mental health services to Bexley, Bromley and Greenwich. We identified lithium monitoring as an area where practice needed to improve through participation in the Prescribing Observatory for Mental Health (POMH) programme.

This programme is coordinated in Oxleas through the local POMH team, which at the time of the baseline audit (2008) comprised representatives from clinical audit, pharmacy, medicine, nursing, and service users. The service user representatives were keen that we should participate in this topic due to their own personal experiences. One had been treated for lithium toxicity in intensive care in the past (as a consequence of not having his treatment regularly monitored), and he also knew people on lithium from local mental health charity groups, who were not having regular blood tests. The other service user said he always had to remind his doctors to take bloods for lithium monitoring and that there didn't seem to be a system in place to flag up when his tests were due.

In most of our community mental health teams (CMHTs), patients would have their prescriptions for lithium written by their GP, who it was assumed would also order the required blood tests and share these results with the psychiatrist who may only see the patient once or twice a year. Initially, as there is no electronic prescribing record, it proved difficult to identify all service users within our Trust who were taking lithium, and this varied by borough.

The pathology laboratory that carried out the blood tests for one of the boroughs produced a spreadsheet of all people who had had a lithium level test in the past month and circulated this to local GPs and psychiatrists. In the other boroughs we had to rely on asking the staff members of all the CMHTs to provide lists of their patients who were taking lithium. Access to blood test results also varied by borough, with a common theme that psychiatrists only received the results of blood tests they had ordered, and not any of those ordered by the GP. This led to duplication of blood tests in some patients and a lack of monitoring in others where it was not clear who, if anyone, was doing the monitoring.

The results of the baseline POMH audit identified that there was indeed a need to improve practice as some of our patients did not seem to have had a blood test for a lithium level in the last year, and only 35% had had 4 tests in the past year as recommended by NICE.

How did you implement the project

Local feedback and action planning.
After the baseline audit, the results were fed back to all relevant staff and patient groups through:

  • Governance meetings
  • Clinical team meetings
  • Pharmacy and nursing newsletters
  • Service user forums
  • Training events for local GPs

Discussions led to the development of local action plans.

Overcoming barriers to change
A common problem of sharing information between pathology, and primary and secondary care was identified. The POMH central team identified that this was a national problem, and not long before the re-audit in April 2010, there was a joint NPSA/POMH event to launch an NPSA alert around lithium monitoring which included a statement that there should be an electronic interface between pathology and primary and secondary care to allow the results of lithium monitoring blood tests to be shared easily.

Development of a lithium database
At the launch event, there was a presentation by a clinical team from another Trust that had performed well in the baseline POMH audit. They described their experience of using a local lithium register. Following disappointing results for Oxleas in the POMH re-audit despite Trustwide engagement in attempts to improve practice, the medical director supported a proposal to the Trust board that Oxleas fund development and on-going management of a lithium database.

The lithium database was designed by a psychiatrist using the computer programme Microsoft Access. Doctors and care coordinators can register their patients on the lithium database via email to the database manager. The manager of the database liaises with the local pathology services once a month to obtain lithium results, and certain renal and thyroid function test results for all our patients prescribed lithium. He uploads this information into the database which then identifies all those patients who are due for a lithium blood test. He then emails the doctor and/or care coordinator for each patient due a blood test to remind them it is due.

Population of the lithium database started in April 2011 and was launched in the Trust through all the same channels as those used for the audit feedback, and also with a stand at the Oxleas Nursing Conference in June 2011. It was not fully operational until September 2011, when the supplementary POMH audit was carried out.

Key findings

Monitoring progress:

Progress was monitored through continued participation in the POMH lithium monitoring Quality Improvement Programme. We collected data for the supplementary audits in September 2011 and July 2013. The results can be seen in the Figures attached in the supporting material section.

Improvements seen:

1. Identification of our patients on lithium.
Through liaison with pathology, we are able to identify all our patients taking lithium and record their details in a central location. At the baseline audit clinical staff identified 251 patients currently treated with lithium. Once the database was fully populated using data from pathology, this number rose to 480. Lithium monitoring was sub-optimal at baseline, and although we did not specifically examine patterns of monitoring in those patients whose lithium treatment was identified solely through liaison with pathology when the database was populated, it is likely that this was poorer than in those patients who teams could initially identify. That is, in order to actively monitor treatment you need to know who to monitor. This finding validates the experiences of the two service users who were involved with the Trust POMH team.

2. Improvement to patient care.
The improvement to patient care, specifically patient safety, can be seen by the significant increase in the number of patients who have had regular blood tests for lithium levels, kidney function and thyroid function.

In the most recent audit, July 2013, the number of patients who had had 4 lithium serum level tests in the last year had risen from 35% at baseline to 52%. The number of patients who had had at least 2 lithium tests in the past year had risen from 78% at baseline to 95%, with the all patients in the sample having had at least one test in the past year.
Clinical teams can request reports from the lithium database, to see what monitoring has been carried out by the GP over the past year, and hence avoid under-monitoring or duplication of testing.

With respect to the effectiveness of treatment, at baseline 67% of patients had a lithium level within the therapeutic range (so high enough to be effective but not high enough to lead to toxicity). By 2013 this proportion had increased steadily to 86%.

Key learning points

Without adequate systems in place to support good clinical practice, it can be very difficult to bring about change even when local clinicians acknowledge problems, engage with them and clearly see the need to improve practice.

The response to practice falling short of standards is often to take a general educational approach. We tried this initially but there was little if any change. This was likely because it was not simply lack of knowledge that was the cause of the gap between the NICE recommendations and clinical practice. It is essential to obtain the views of staff involved in the process to understand what the perceived barriers to best practice are and time must be allowed for this to happen. When the intervention that is put in place directly addresses this gap, in this case the perceived lack of an automated reminder system, positive changes in practice can occur. It is known that personalised reminders have a greater impact on practice than general educational initiatives, and this may be a factor in the success of our initiative.

In the case of lithium monitoring specifically, there needs to be a good working relationship between those who analyse the blood samples and those who need to know the results to provide safe and effective care. Given the number of interfaces between clinical teams, it is essential to have good local systems in place.

Contact details

Sarah Elliott
Specialist Clinical Pharmacist for Quality Improvement
Oxleas NHS Foundation Trust

Secondary care
Is the example industry-sponsored in any way?