• Welcome to Phoenix Rising!

    Created in 2008, Phoenix Rising is the largest and oldest forum dedicated to furthering the understanding of and finding treatments for complex chronic illnesses such as chronic fatigue syndrome (ME/CFS), fibromyalgia (FM), long COVID, postural orthostatic tachycardia syndrome (POTS), mast cell activation syndrome (MCAS), and allied diseases.

    To become a member, simply click the Register button at the top right.

Collin, Crawley, White: Trends in the incidence of CFS and FM in the UK

mango

Senior Member
Messages
905
Trends in the incidence of chronic fatigue syndrome and fibromyalgia in the UK, 2001-2013: a Clinical Practice Research Datalink study

Collin SM
1, Bakken IJ2, Nazareth I3, Crawley E1, White PD4.

Author information
  1. School of Social and Community Medicine, University of Bristol, Bristol BS8 2BN, UK.
  2. Norwegian Institute of Public Health, 0403 Oslo, Norway.
  3. UCL Department of Primary Care and Population Health, UCL Royal Free Campus, London NW3 2PF, UK.
  4. Wolfson Institute of Preventive Medicine, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London EC1M 6BQ, UK.
J R Soc Med. 2017 Jan 1:141076817702530. doi: 10.1177/0141076817702530. [Epub ahead of print]

Abstract
Objective
Trends in recorded diagnoses of chronic fatigue syndrome (CFS, also known as 'myalgic encephalomyelitis' (ME)) and fibromyalgia (FM) in the UK were last reported more than ten years ago, for the period 1990-2001. Our aim was to analyse trends in incident diagnoses of CFS/ME and FM for the period 2001-2013, and to investigate whether incidence might vary by index of multiple deprivation (IMD) score.

Design
Electronic health records cohort study.

Setting
NHS primary care practices in the UK.

Participants
Patients registered with general practices linked to the Clinical Practice Research Datalink (CPRD) primary care database from January 2001 to December 2013.

Main outcome measure
Incidence of CFS/ME, FM, post-viral fatigue syndrome (PVFS), and asthenia/debility.

Results
The overall annual incidence of recorded cases of CFS/ME was 14.8 (95% CI 14.5, 15.1) per 100,000 people. Overall annual incidence per 100,000 people for FM was 33.3 (32.8-33.8), for PVFS 12.2 (11.9, 12.5), and for asthenia/debility 7.0 (6.8, 7.2).

Annual incidence rates for CFS/ME diagnoses decreased from 17.5 (16.1, 18.9) in 2001 to 12.6 (11.5, 13.8) in 2013 (annual percent change -2.8% (-3.6%, -2.0%)). Annual incidence rates for FM diagnoses decreased from 32.3 (30.4, 34.3) to 27.1 (25.5, 28.6) in 2007, then increased to 38.2 (36.3, 40.1) per 100,000 people in 2013.

Overall annual incidence of recorded fatigue symptoms was 2246 (2242, 2250) per 100,000 people. Compared with the least deprived IMD quintile, incidence of CFS/ME in the most deprived quintile was 39% lower (incidence rate ratio (IRR) 0.61 (0.50, 0.75)), whereas rates of FM were 40% higher (IRR 1.40 (0.95, 2.06)).

Conclusion
These analyses suggest a gradual decline in recorded diagnoses of CFS/ME since 2001, and an increase in diagnoses of fibromyalgia, with opposing socioeconomic patterns of lower rates of CFS/ME diagnoses in the poorest areas compared with higher rates of FM diagnoses.

KEYWORDS:
Chronic fatigue syndrome; diagnosis; fibromyalgia; general practice; incidence; fatigue; myalgic encephalomyelitis; post-viral fatigue; primary care

https://www.ncbi.nlm.nih.gov/pubmed/28358988

http://journals.sagepub.com/doi/10.1177/0141076817702530
 

Dolphin

Senior Member
Messages
17,567
We note that the observed annual incidence for patients <20 years of age (10 per 100,000) translates into a prevalence of 1–2% (if we assume an average 12–24 months’ duration of illness), which is entirely consistent with paediatric CFS/ME prevalence estimates from population-based studies.31
This doesn't make sense to me. 1% is 1000 per 100,000 which is 100 times the incidence of 10 per 100,000.
 

Dolphin

Senior Member
Messages
17,567
However, our estimated incidence of new CFS/ME diagnoses for the period 2008–2010 (in 2009, 14.5 (95% CI 13.6 to 15.8) per 100,000) is not inconsistent with estimates based on data from NHS specialist CFS/ME services in England (in 2009, 22 (95% CI 17 to 29) per 100,000), if we consider that the latter figure includes re-referrals.14 Incidence of CFS/ME diagnoses in 2009 from our data, including patients who had a previously recorded CFS/ME diagnosis, is 17.0 (95% CI 15.8 to 18.3). This consistency with estimates from clinical services would suggest that the majority of patients who have a CFS/ME diagnosis recorded by their GP have had this diagnosis made or confirmed by a specialist service. Our overall annual incidence rate estimate of 14.8 (95% CI 14.5 to 15.1) per 100,000 is also consistent with a rate of 15 (95% CI 6 to 41) per 100,000 reported for the period 2007–2010 based on data from a primary care cross-sectional study in three English regions.25
 

Dolphin

Senior Member
Messages
17,567
We would expect that a substantial proportion (if not all) PVFS cases should have been classified as CFS/ME. Viral infections are known to trigger CFS/ME,32 and fatigue of ≥4 months’ duration after the acute phase of an infection has passed should, according to NICE criteria, warrant consideration for a diagnosis of CFS/ME.8 Towards the end of our study period, this reclassification would increase the incidence of CFS/ME by around 50%
 

Dolphin

Senior Member
Messages
17,567
Population-wide incidence of CFS/ME in Norway from 2008 to 2012 as indicated by ICD-10 code G93.3 (‘post-viral fatigue syndrome/benign myalgic encephalomyelitis’) was 25.8 (25.2, 26.5) per 100,000 person years.33 This is 74% higher than the incidence in our study, possibly because the ICD-10 code combines PVFS and CFS/ME in a single classification.
 

user9876

Senior Member
Messages
4,556
This doesn't make sense to me. 1% is 1000 per 100,000 which is 100 times the incidence of 10 per 100,000.

I would assume that these are new diagnoses each year and hence the overall prevalence will be higher although I don't know how they calculate that.
 

Dolphin

Senior Member
Messages
17,567
I would assume that these are new diagnoses each year and hence the overall prevalence will be higher although I don't know how they calculate that.

It is right that incidence should not be confused with prevalence. But I don't see how that explains the figures in this:

We note that the observed annual incidence for patients <20 years of age (10 per 100,000) translates into a prevalence of 1–2% (if we assume an average 12–24 months’ duration of illness), which is entirely consistent with paediatric CFS/ME prevalence estimates from population-based studies.31
 

Snow Leopard

Hibernating
Messages
5,902
Location
South Australia
I agree this looks completely wrong. I cannot quite be sure of the sums at the moment but I cannot see how they can be the way they say.

I think the prevalence should be 0.01-0.02% - which is actually far too low.

I believe the incidences are reported in the usual person/years (haven't looked closely).

If a majority aren't recovering, then it adds up to somewhere between 0.1-1% (lifelong illness, depending on assumptions used).

But these aren't true incidences, they're number of diagnoses logged in the system. If medical doctors aren't doing their jobs properly and making sure patients have diagnoses and that these diagnoses are logged into the system, then will get biased results.
 
Last edited:

Snow Leopard

Hibernating
Messages
5,902
Location
South Australia
It is right that incidence should not be confused with prevalence. But I don't see how that explains the figures in this:

Strange that they didn't bother to cite Jason, Bell etc.

Crude assumptions (10 per 100000 * 20 years) / 2 years duration = 0.1% (this math also assumes that once a person is ill, and recovered, they can become ill again), hmm...

edit - Note, I'm assuming the incidences are in person/years

Figure 4 of incident vs age is interesting
http://journals.sagepub.com/doi/10.1177/0141076817702530#
in that it doesn't reproduce the double hump that was in a Norwegian study

Would have liked to see it broken down into smaller age ranges. It is mistake to pool the 0-20 in one group in comparison to the adults.
 
Last edited:

user9876

Senior Member
Messages
4,556
If a majority aren't recovering, then it adds up to somewhere between 0.1-1% (lifelong illness, depending on assumptions used).

But these aren't true incidences, they're number of diagnoses logged in the system. If medical doctors aren't doing their jobs properly and making sure patients have diagnoses and that these diagnoses are logged into the system, then will get biased results.

They also require 12 months recording of data from the GP so that there is not an effect where new patients get a diagnosis. I think this could have an affect on the figures as sometimes it can be necessary to move GPs to get anyone who can be bothered to look for a diagnosis.
 

Dolphin

Senior Member
Messages
17,567
If a majority aren't recovering, then it adds up to somewhere between 0.1-1% (lifelong illness, depending on assumptions used).
Though not as teenagers as claimed by this study as there not many years to accumulate. Also they claim the diagnosis only last 12 to 24 months in teenagers.

Anyway it's only a minor point.

Edited to add: this was written before I saw your latest post.
 

Snow Leopard

Hibernating
Messages
5,902
Location
South Australia
Though not as teenagers as claimed by this study as there not many years to accumulate. Also they claim the diagnosis only last 12 to 24 months in teenagers.

Initial onset is often relapse-remission for teenagers, later to develop into continuous illness for many, and this is something that is not effectively captured with the methodology of this study. The data is what it is, just important not to jump to conclusions about duration with such limitations.
 

Dolphin

Senior Member
Messages
17,567
Incidence = number of new cases each year divided by population.

Prevalence = number of all cases divided by population.
I know that. I don't see how that explains these figures:

We note that the observed annual incidence for patients <20 years of age (10 per 100,000) translates into a prevalence of 1–2% (if we assume an average 12–24 months’ duration of illness), which is entirely consistent with paediatric CFS/ME prevalence estimates from population-based studies.31