The impact of a county-wide lithium monitoring database

What is lithium?
In the 5th century AD a Roman physician was reputed to have recommended the use of alkaline waters to patients suffering with mental disorders. The effect of these waters on their mental state is thought to be due to the lithium levels found in them (Marmol 2008). Since the 1800s lithium has been used to treat recurrent depression and other depressive states. It is now the ’Gold standard‘ treatment for bipolar disorder (manic-depression), both for mania and relapse prevention. Lithium is also used with other antidepressant treatments in difficult to treat depression (Cade 1949; Collins, Barnes et al. 2010).

Why does lithium need monitoring?
Lithium can be a highly effective treatment but is known to have many side effects, some of which can be so troubling that treatment has to be stopped. Most of these side effects are ’dose-related‘ i.e. the higher the blood level of lithium, the worse the side effect (Schou 1999). Lithium has a ’therapeutic window‘ in that too low a blood level and it won’t work, too high a dose and you get side effects. In Europe the recommended usual range for lithium levels in the blood is 0.6-0.8mmol/L, although the dose and therefore blood level should be adjusted in order to get the best outcome for each patient (Schou 1999; NICE 2006). Lithium is usually not thought to be effective with levels lower than 0.4mmol/L, and very few people will benefit from levels of more than 1.0mmol/L (Collins, Barnes et al. 2010). Because of this, all treatment guidelines suggest that lithium levels should be tested regularly while the person is taking lithium.

The body removes lithium via the kidneys so if the kidneys are not working properly this can lead to less lithium being excreted. This can lead to increasing lithium levels and an increase in side effects (Collins, Barnes et al. 2010). Drinking more or less water than normal or sweating a lot (for example in a hot climate) can also alter the levels of lithium making it either ineffective or into the toxic range. Other medicines such as NSAIDs (used for arthritis and pain e.g. ibuprofen) and diuretics (water tablets) can also raise lithium levels. The NICE guidelines on bipolar disorder recommend that, before starting treatment with lithium, kidney and thyroid function tests should be carried out. These guidelines then recommend that these tests should be repeated every 6 months, with lithium levels checked every 3 months (NICE 2006). The British Association for Psychopharmacology (BAP) guidelines recommend that kidney and thyroid function are only tested every 12 months, with lithium levels checked every 3-6 months in people on a stable dose. BAP also states that these tests should be carried out more often if the physical state of the patient or their symptoms change (BAP 2009).

In the UK the Quality and Outcomes Framework (QOF) also sets targets for GPs (in addition to NICE) for monitoring people on lithium in the community (NICE 2012). These targets are linked to payments for services and only require GPs to report the percentage of people taking lithium who have had their thyroid and kidney function checked in the last 15 months, and those who have had a lithium level within the recommended range within the last 6 months (Collins, Barnes et al. 2010). This is fine if you are one of the people who have had all the tests done properly. We, the authors of this article, think that the target should be 100% of people getting the right monitoring.

How is the UK doing with meeting these monitoring targets?
Despite these clear guidelines on how often lithium levels should be checked, surveys in the UK over the last 20 years have found that many services haven’t been managing to meet these rates of monitoring (Collins, Barnes et al. 2010). In 2009, 38 Mental Health Trusts from across the UK took part in a survey on lithium monitoring, which was run by the Prescribing Observatory for Mental Health (POMH-UK). Trusts sent in data for a total of 2976 people who had been on lithium for more than a year (Collins, Barnes et al. 2010).


publicinfo-lithium-fig1If monitoring was completed as per NICE guidelines, people should have at least two renal (kidney) and thyroid function tests recorded and at least three lithium level tests per year. From the graph above it can be seen that less than 1 in 3 (30%) of people had lithium tests meeting NICE standards, only 1 in 2 (49%) for thyroid function tests and just over 1 in 2 (56%) for renal function tests. This is well below the standards set and leaves many people at risk of serious and avoidable side effects.

Implementation of the Norfolk database
In May 2000 a pharmacy-led prescribing group came up with the idea of a Norfolk-wide lithium register and database. This came about after long-term concerns over a lack of a consistency in monitoring and a lack of a shared care agreement across the county e.g. if a lithium level was too high it was unclear whether the GP should take action or a psychiatrist. Sometimes both contacted the patient, sometimes neither did.
The initial aim of the database was to provide a service both to clinicians and patients to make sure that all people on lithium were given proper information, education and specialist advice and receive regular blood tests according to the guidelines (Holmes 2005) and a shared care agreement. A successful shared care agreement allows for a smooth transfer of someone’s treatment from secondary care (hospitals) to GPs (DoH 2007). The Norfolk lithium database started operation in May 2002, and the computer system was upgraded in 2011 and is now known as System TDM®.  As soon as the lithium test is reported by one of the labs across Norfolk it gets sent to System TDM® and the patient starts the registration process if they are not already on the system.

Impact of the lithium database on rates of lithium testing in Norfolk
The database has been going for ten years in Norfolk and the data now shows the impact of a proper organised system. All those people who had lithium, thyroid or renal function tests recorded on the database between June 2002 and June 2012 were included in the analysis. The data includes people who had been on lithium for less than one year or who stopped so they would not have had the recommended number of tests in a year.


publicinfo-lithium-fig2As you can see there has been a noticable increase in the number of lithium tests carried out for each person. From the graph above you can see that:

  • Over 2 in 3 (68.5%) people had at least 3 lithium tests
  • Nearly 8 of 9 (87.6%) had 2 or more thyroid function tests
  • Over 5 in 6 (85.3%) had 2 or more renal function tests

What does this show?
Since the original database was started in Norfolk in 2002 there has been a steady increase in the number of people receiving the recommended number of lithium, renal and thyroid function tests. In December 2009 the National Patient Safety Agency (NPSA) issued a ’patient safety alert‘ requiring action by all health care organisations in UK to improve the safety of lithium therapy (NPSA 2009).  In the five years before this NPSA alert, 560 ’patient safety incidents‘ had been reported in people taking lithium. The key to all these reports was lack of lithium monitoring (NPSA 2009). For comparison in Norfolk over the same period of time there had been no reports related to lithium (Cree 2011). There is thus a clear benefit for setting up a local lithium database to help meet NICE and BAP targets and thus vastly reducing the risk of side effects from lithium. It also means that lithium, which is generally accepted as being the most effective treatment for bipolar disorder, can be prescribed safely. Lithium seems to be prescribed in Norfolk about twice as often as other areas of the UK, and it may be that lithium is not being prescribed for people in other areas who would benefit from it due to concerns on safety and monitoring.

What next?
With ten years data on all 1500 people taking lithium in Norfolk, we are now researching what is a unique set of data. There are still many unanswered questions. Does kidney or thyroid damage occur with lithium in the usual range? What happens if you have an occasional high level? Should levels be checked more often or is 3-6 months too often? We are also extending System TDM to Suffolk which will allow us to compare a ’before and after‘ group of people. Come back in a few years and see what we’ve found out!

BAP (2009). Evidence-based guidelines for treating bipolar disorder: revised second edition—recommendations from the British Association for Psychopharmacology.
Cade, J. (1949). “Lithium Salts in the Treatment of Psychotic Excitement.” The Medical Journal of Australia 2(10): 349-520.
Collins, N., T. R. Barnes, et al. (2010). “Standards of lithium monitoring in mental health Ttrusts in the UK.” BMC Psychiatry 10: 80-80.
Cree, N. (2011). “Why patients on lithium therapy get a safer deal if they are based in Norfolk.” Pharmaceutical Journal 286: 170.
DoH. (2007). “Department of Health: Shared care guidelines.” from .
Holmes, B. K. (2005). Lithium Monitoring: The Patient Perspective. School of Nursing and Midwifery. Norwich, University of East Anglia. Master of Science by Research.
Marmol, F. (2008). “Lithium: Bipolar disorder and neurodegenerative diseases Possible cellular mechanisms of the therapeutic effects of lithium.” Progress in Neuro-Psychopharmacology & Biological Psychiatry 32(8): 1761-1771.
NICE (2006). National Institute for Clinical Excellence.Bipolar disorder: The management of bipolar disorder in adults, children and adolescents, in primary and secondary care. Clinical Guideline 38 2006.
NICE. (2012). “About the Quality and Outcomes Framework (QOF).” from .
NPSA (2009). “Safer Lithium Therapy.”
Schou, M. (1999). “Lithium treatment at 52.” Journal of Affective Disorders 67: 21-32.

Leave a Reply

Your email address will not be published. Required fields are marked *