Yesterday, Dutch ZonMw announced which of the three consortia and 22 research grant applications were allotted which amounts. Summaries of the projects also were published on ZonMw’s site.
As one of the summaries indicates, the aim of ME/CFS Lines is to build a longitudinal ME/CFS patient cohort and biobank based on the existing population cohort Lifelines, in which ME/CFS diagnostic criteria have been applied since 2014.
Lifelines contains the data of 167,000 people. According to the Lifelines website, those data are representative of the whole population: (…) These findings indicate that the risk of selection bias is low and that risk estimates in Lifelines can be generalised to the general population.
In 2018, Lifelines included data and biomaterials from 2,500 participants with ME/CFS; since then, another 400 participants have developed ME/CFS.
This does raise some questions.
1. According to the 2005 Dutch Health Council report, the number of people with ME/CFS in the Netherlands was estimated at 30,000 to 40,000. The Health Council committee in 2018, which included Professor Rosmalen, saw no reason to arrive at a different estimate[1]
.
A systematic review by Lim and others in 2020 came to a prevalence of 0.89% (including those younger than 18 years) if the CDC 94 criteria are used. These are very broad criteria. The CDC discarded their criteria and replaced them with the more stringent IOM criteria. Converted to the entire Dutch population, that would come out to 158,420 patients, about four times the Dutch Health Council’s estimate
Looking at the people identified as ME/CFS patients in Lifelines were the participants are 18 years or older, it is notable that in 2018 the number of people with ME/CFS was 2,500, and in the next scarcely four years another 400 were added. So that’s a total of 2,900 people suffering with ME/CFS out of a population of 167,000. That would mean a prevalence of 1.74% in the adult population.
For the whole of the Netherlands[2], that would translate to 254.040 adults suffering with ME/CFS. Almost twice as many as Lim and his colleagues had calculated for the whole population, and 6 to 8 times as many as The Health Council. We know from a survey[3] that 24% of the Dutch ME/CFS patiënts mentioned they were suffering with ME/CFS before the age of 20 years. When we take that into account the prevalence would raise to over 300.000.
That’s weird.
And then there is another problem:
2. Lifelines claims that data from Lifelines can be generalised for the general population
Our results suggest that, adjusted for differences in demographic composition, the LifeLines adult study population is broadly representative for the adult population of the north of the Netherlands. The recruitment strategy had a minor effect on the level of representativeness. These findings indicate that the risk of selection bias is low and that risk estimates in LifeLines can be generalised to the general population.
But this claim does not hold true for ME/CFS patients. After all, severe and very severe ME/CFS patients are house- or bed-bound and hardly ever visit doctors or hospitals. Those data are missing from Lifelines and therefore also from ME/CFS Lines.
Overall, the estimate is that about 25% of patients with ME/CFS fall into this severe category. This would mean that Professor Rosmalen’s numbers include only 75% of patients. If we correct Professor Rosmalen’s numbers that would mean there are 1.33 x 254,040 = 338,000 adult patients suffering with ME/CFS in the Netherlands. That is about 10 times the Health Council’s estimate, all ages included, while the Health Council committee she was a member of in 2018, saw no reason to adjust the estimate.
What could be going on here?
a. It makes perfect sense. We have 10 times more patients with ME/CFS than was thought.
In that case we are dealing with a world-class discovery even before the studies actually started.
How can it be that we didn’t already know this?
Why hasn’t Professor Rosmalen published that discovery yet?
There are also other options:
b. Lifelines as a whole is not at all as representative as claimed.
c. In any case, it can be established that Lifelines is not representative when it comes to ME/CFS, because the group of seriously to very seriously ill patients is missing.
d. The method of selecting patients with ME/CFS is flawed and leads to far too large a selection
What criteria were applied?
What criteria were applied? The question is also whether a doctor made the diagnosis or is it just an assumption based on data from Lifelines? It is a well-known fact that selection based on self-reporting leads to a much larger patient group than if the diagnosis is made by a doctor. This is also laid down in the Health Council’s advice Professor Rosmalen was a member of the ME/CFS advisory committee and must therefore be familiar with its contents.
Careless selection of ME/CFS patients leads to biased research findings. This has been established by the US CDC, among others, and by the also US Agency for Health and Care Research Quality.
This is a very big problem!
Impressive studies can be done, but if the patient group is too broad, important common features are not found because the patients are too different from each other. Then we end up in the situation where somatically insufficient explanation is found for the physical complaints (SOLK/MUS). The lack of sufficient explanation is then a consequence of a seriously flawed study base don a far too broad (heterogeneous) study population. The findings of the biomedical explanation will vanish in this mass or is it mess.
For that reason, the text of the ME/CFS biomedical research programme includes the requirement:
“In building the biobank and patient registration, much attention will have to be paid to the quality of patient selection. This quality should be ensured by following an established diagnostic protocol and assessment performed by expert and experienced clinicians. This may require linking up with expertise abroad.”
Conclusion: the ME/CFS-Lines consortium seems to fall short by far of ZonMw’s requirements.
- How could it be possible that a 4,4 million-euro grant was awarded to ME/CFS-Lines and its sub-studies?
- How could the applications of this consortium and the substudies pass ZonMw’s pre-selection?
- What was the advice of the four patient representatives involved in this ZonMw programme? After all, they had weighty advisory rights, after being deprived of their voting rights by ZonMw for yet unclear reasons.
- What did the programme committee do with that advice?
Can ZonMw explain how this debacle could have happened?
For sure the last word hasn’t been said or published about this matter. Follow-up actions are being planned.
So keep an eye at this site.
Rob Wijbenga
editor-in-chief
[1] In 2005, the Health Council provided an overview of the already wide variety of prevalence figures available at the time and, based on this, expressed the suspicion that there were then between 30,000 and 40,000 patients with ME/CFS in the Netherlands, the majority (75-80%) of whom were women. Based on the research published since then, the committee saw no reason to adjust this estimate which amounts to about 180 to 250 patients per 100,000 (wich means 32,500 to 44,500 patients in 2022).
[2] The population of the Netherlands reached 17.8 million inhabitants by the end of 2022. 3.2 million are younger than 18 years. Source CBS: Population counter (www.cbs.nl )
The ME Global Chronicle 