Public Health England estimates that there could be up to 3,000 new cases of Lyme disease per year in the UK whereas Caudwell LymeCo charity and Lyme Disease UK estimate that the true number could be around 45,000. Here is the explanation of the calculations behind the estimate:
Reported numbers are a fraction of the real total
It is widely accepted that the number of UK cases of Lyme disease recorded centrally, by Public Health England and Health Protection Scotland, is only a fraction of the total.
Why is this?
Three reasons for under-reporting of Lyme cases
1 – False negative blood tests at national reference laboratories
Firstly, there are the false negative blood tests. Let’s use P to represent the total number of people recorded each year as having Lyme disease. P people have their blood sent to a national reference laboratory, and the test results are positive.
There is no such thing as a blood test which gets the right result every time, however. So let’s call all the blood samples which are really positive, but which get a negative test result, FN.
This means the total positive blood samples sent to the national reference laboratories = P + FN. In words, this means the samples with a positive result, plus the samples with a false negative result.
P + FN
2 – False negative blood tests at local hospitals
This total, P+FN, is still not the real total of positive blood samples.
Before patients’ blood is sent to one of the national reference laboratories, it is tested at a local hospital using an ELISA test. It is only sent to the national reference laboratories if this first ELISA test is positive. This ELISA test also gets some false negatives, too.
So, we already know the total of positive blood samples sent to the national reference laboratories is P + FN. These are the ones which the local testing labs found positive.
But what about the samples which the local testing labs mistakenly found negative? Let’s call these regionally tested false negative samples RFN.
This means that, of all the blood samples sent off by NHS doctors for testing, the grand total of all the positive ones = P + FN + RFN.
However, the only ones from these who are recorded in national statistics as having Lyme disease are P.
P + FN + RFN
3 – Patients whose blood is never tested
The UK national guidelines, produced by Public Health England, are clear that an erythema migrans circular rash is diagnostic of Lyme disease. This means that, if doctors see this unique rash on a patient and identify their illness as Lyme disease, the diagnosis is confirmed and they should not send a blood sample off for testing.
Let’s use R to represent these patients who are diagnosed by rash rather than blood test.
Now we can add up to total of all Lyme disease patients in the UK. This total equals P + FN + RFN + R.
P + FN + RFN + R
What are the real numbers?
Now, we need to start finding out the numbers to calculate the real total.
If we look in the medical literature, we can find independent assessments of the sensitivity and specificity of the diagnostic tests. We can also find statements by medical experts regarding the percentage of patients who have an erythema migrans circular rash.
All of the source documents and official data published online is referenced in a presentation (link below) which explains our methodology for this calculation in detail.
0 – What was P?
First, we can easily state that P was 954 in 2014. This is the total number of Lyme disease cases confirmed by blood testing by the two national reference laboratories in the UK.
We will use this as our base number for all the following calculations.
P = 954
1 – How much is FN?
We have to calculate FN in two stages, because the national reference laboratories use two tests. They consider samples positive if they are positive in the both the first test (called an ELISA test) and the second test (called a western blot).
We took the percentage of false negatives found in an independent assessment of each of the tests which was posted on the test kit manufacturers’ websites.
At the national reference laboratory for England, Wales and Ireland, called RIPL, the false negatives found by using first the ELISA and then the western blot are 77% of the total blood samples tested. At Raigmore, the national reference laboratory for Scotland, they are 70%.
When we add up these totals, FN = 2,945. Therefore…
P + FN = 3,899
In other words, the number of people represented by P + FN is about the same as two regiments of the British army.
2 – How much is RFN?
The false negatives from the regional laboratories are 45% of all samples tested. This means that:
RFN = 3,177
If all of these blood samples, found negative by mistake, were found positive, then the total number of positive samples would have been 7,076 in 2014.
P + FN + RFN = 7,076
3 – How many patients are in the R category?
Now we need to add on the number of patients in category R, whose blood is never tested because they can be diagnosed by a distinctive erythema migrans rash instead.
Nobody gathers data on this objectively, so we have to base our calculations on surveys of sample patient populations, or estimates issued by reliable authorities. We have chosen to use three UK sources which we think are likely to have good credibility, and to avoid using sources from the United States or other countries where different strains of borrelia are prevalent and the total could be different from the UK.
- The British Infection Association says 90% of Lyme disease patients have erythema migrans.
- Public Health England estimates that two thirds – that is, 66% – of patients have erythema migrans.
- Health Protection Scotland, in a research paper based on a patient survey by Mavin et al, estimate that 48% of Lyme patients had an erythema migrans.
These numbers clearly vary significantly, so we have performed the same calculation using each of them. So, what was P + FN + RFN + R, the grand total number of Lyme patients in the UK?
- If we believe the British Infection Association on the incidence rate of erythema migrans in UK Lyme patients, there were 70,760 UK Lyme cases in 2014.
- If we believe Public Health England on the incidence rate of erythema migrans in UK Lyme patients, there were 20,812 UK Lyme cases in 2014.
- If we believe the Health Protection Scotland research paper by Mavin at al on the incidence rate of erythema migrans in UK Lyme patients, there were 13,634 UK Lyme cases in 2014.
The fact that these numbers vary so dramatically is frustrating, particularly since we have no way of evaluating which of them is likely to be the most reliable. Given this, we have chosen to give them each equal credibility and we therefore assume the median figure.
This means that we estimate there are 45,000 new cases of Lyme disease in the UK each year.
What does this figure really mean?
45,000 people are enough to fill Stratford Olympic stadium, or the O2 arena twice, or 100 jumbo jets.
With the current UK life expectancy of 81.5 years, the numbers imply that 1 person in 20 will catch Lyme disease at some point in their life.
This means in every school class in England, there is at least one child who will catch Lyme disease at some point.
Sources and further reading
- Public Health England Guidance: Lyme borreliosis epidemiology and surveillance, Published 1 May 2013, https://www.gov.uk/government/publications/lyme-borreliosis-epidemiology/lyme-borreliosis-epidemiology-and-surveillance
- Distribution and presentation of Lyme borreliosis in Scotland – analysis of data from a national testing laboratory, S Mavin, EJ Watson, R Evans, J R Coll Physicians Edinb 2015 45: 196–200, https://www.rcpe.ac.uk/sites/default/files/jrcpe_45_3_mavin.pdf
- Mehnert WH, Krause G. Surveillance of Lyme borreliosis in Germany, 2002 and 2003. Euro Surveill 2005; 10: 83–5.
- Christova I, Komitova R. Clinical and epidemiological features of Lyme borreliosis in Bulgaria. Wien KlinWochenschr 2004; 116: 42–6.
- Bacon RM, Kugeler KJ, Griffith KS et al. Lyme disease – United States, 2003-2005. MMWR 2007; 56: 573–6.
- Lehman PV et al.: Unique Strengths of EliSpot for T Cell Diagnostics in: Kalyuzhny AE. Handbook of EliSpot: Methods and Protocols, Methods in Molecular Biology, Vol. 792. 2nd Ed: Springer; 2012: 3-23
- Von Baehr et al.: Evaluation of a New Multiparametric Microspot Array for Serodiagnosis of Lyme Borreliosis in: Clin. Lab. 2015;61
- Chenggang Jin et al.: An enhanced ELISPOT assay for sensitive detection of antigen specific T cells responses to Borrelia burgdorferi, Cells 2013, 2, 607-620; doi 10.3390/cells2030607
- Large differences between test strategies for the detection of anti-Borrelia antibodies are revealed by comparing eight ELISAs and five immunoblots, C. W. Ang, corresponding author D. W. Notermans, M. Hommes, A. M. Simoons-Smit, and T. Herremans
- Evaluation of two line immunoassays for the detection of Borrelia antibodies, Verstreken I., Appeltans T., Verhaegen J., Lagrou K.
- Borrelia burgdorferi VlsE antigen for the serological diagnosis of Lyme borreliosis, Marangoni A1, Moroni A, Accardo S, Cevenini R.
- Laboratory diagnosis of Lyme neuroborreliosis: a comparison of three CSF anti-Borrelia antibody assays, A. J. Henningsson, M. Christiansson, I. Tjernberg, S. Löfgren, and A. Matussek
- The epidemiology, prevention, investigation and treatment of Lyme borreliosis in United Kingdom patients: a position statement by the British Infection Association, 2011, British Infection Association
- Lyme borreliosis epidemiology and surveillance, Public Health England, Published 1 May 2013 https://www.gov.uk/government/publications/lyme-borreliosis-epidemiology/lyme-borreliosis-epidemiology-and-surveillance
- Lyme Borreliosis Information leaflet, European Centre for Disease Control http://ecdc.europa.eu/en/healthtopics/vectors/world-health-day-2014/Documents/factsheet-lyme-borreliosis.pdf
- Treatment Guidelines: Evidence Assessments and Guideline Recommendations in Lyme disease:The Clinical Management of Known Tick Bites, Erythema Migrans Rashes and Persistent Diseaseby The International Lyme and Associated Disease Society (ILADS)(http://www.ilads.org/lyme/treatment-guideline.php#sthash.OdERgdpL.dpuf)
- 2-tiered antibody testing for early and late Lyme disease using only an immunoglobulin G blot with the addition of a VlsE band as the second-tier test, by Branda JA, Aguero-Rosenfeld ME, Ferraro MJ, Johnson BJ, Wormser GP, Steere AC.
- Post-treatment Lyme disease syndrome symptomatology and the impact on life functioning: is there something here? Aucott JN1, Rebman AW, Crowder LA, Kortte KB.
- Infection report Volume 9 Number 41 Published on: 20 November 2015 Zoonoses Common animal associated infections quarterly report (England and Wales): third quarter 2015; and Lyme disease 2013-2014 data downloadable at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/478807/hpr4115_zoos.pdf viewed on 29/06/2016
- Statistical bulletin: Life Expectancy at Birth and at Age 65 by Local Areas in England and Wales: 2012 to 2014, Office of National Statistics (viewable at ons.gov.uk)