ME/CFS Epidemiology - sex ratios, female predominance

Posts moved from:
Suppose you have €5-10 million for ME/CFS research, what would you spend it on?

I haven't collected up all the evidence in one place, but my impression has been of an incidence of more like two thirds female, one third male. Regardless, here are some possible ideas of the top of my head as to why reported rates might not be accurate:

• Bias in exposure to ME/CFS triggers (or factors that make triggers more potent) in certain studies that aren't reflective of overall incidence. Cultural bias in what infections/troggers are regarded as causing ME/CFS. In the case of Q-fever fatigue syndrome (and possibly Ross River fever too), I think there might even be more males than females, because of the gender patterns in exposure to the pathogens. Some people discount QFFS as ME/CFS, saying that the people who get QFFS are 'made of different stuff' to those who get ME/CFS. There was a suggestion that more young females were exposed to the giardia outbreak in Norway due to them being more likely to drink lots of water. I think a lot of Gulf War Illness is in fact ME/CFS, but it doesn't get diagnosed as that.

• I believe, from observing young people with ME/CFS, that males are less likely to admit there is a problem if the disease is mild than females are. They may prefer to be viewed as lazy than labelled with a stigmatised 'it's all in your mind' disease. That might be why some people have concluded that the disease tends to be more severe in males - it's just the most affected who get the label. Also, I think culture has (at least in the past) tended to allow males with a mild ME/CFS disease to continue to function without a disease label - they may continue to work and then come home and slump in a chair. There hasn't been the same expectation that they will come home and get dinner on and look after the children. I think it's generally accepted that women are more likely to go to the doctor with a health issue than men - that may not be true and certainly won't apply across the board, but it is a prevailing stereotype that may have some basis in fact.

• Doctors may be more mindful of the stigma that goes with a CFS diagnosis when they have a male patient, and spend more time seeking alternative diagnoses, delaying or avoiding the CFS diagnosis. They may be less worried about that for females, in fact they may assume that CFS is a woman's disease, and so may tend to readily reach for a CFS diagnosis when a female patient sits in front of them complaining of fatigue.

• Males tend to have higher incidences (or at least diagnoses) of some other "serious" diseases, with biomarkers, that can cause fatigue, so ME/CFS symptoms may be attributed to those other diseases. Females may have lifestyles that are more likely to cause fatigue (working mother with inadequate help for housework, breastfeeding, anaemia due to blood loss) and so phone surveys supposedly screening for CFS that ask things like 'have you been fatigued for more than 6 months while not having a medical condition that explains the fatigue' may skew towards females.

• Further on the point that having the symptoms of ME/CFS may be harder to accept and less accepted in some cultures if you are a male: This may lead to higher rates of self-medication with alcohol and other drugs and perhaps even suicide in men than in women. I suspect that some of the people who end up homeless have undiagnosed ME/CFS - they just don't have the energy to do what is required to get themselves out of that situation, and family and friends have given up on them.

 

From a discussion on another thread about a claimed sex ratio in ME/CFS of 5 to 1:

I know the issue of sex ratios issue isn't very important in a paper about biomarkers, but there are good reasons to try to get it as right as possible in published papers. I'm a bit surprised about the belief that the voluntary DecodeME participation rates is the best data on sex ratios around, resulting in the use of a 5 to 1 sex ratio. DecodeME required that people had had a medical diagnosis of ME/CFS, and then it is highly likely that there was some bias skewing towards female participation in the study overlaid on that. In almost any voluntary study open to both sexes that doesn't actively recruit for balanced representation, there is more female participation. I think the 5 to 1 sex ratio is out of step with the ratio reported in most literature I have seen, which tends to use a 3 to 1 ratio. A 3 to 1 ratio is in fact the 'old ground'; a claim of a 5 to 1 ratio is a departure that I think needs to be well-evidenced.

If the ICD10 G93.3 codes in England suggest a ratio of 4 to 1, wouldn't that be a more solid number to tie a claimed sex ratio to than the DecodeME participation rates? Arguably, the stigma of an ME/CFS diagnosis is particularly high in the UK, and that may affect male identification with, and participation in, the ME/CFS community. Ongoing claims that ME/CFS is a 'female disease' may actually exacerbate the stigma some men feel with the diagnosis, and also contribute to lower rates of diagnosis in men (in a similar way to past under-diagnosis of heart disease in females). Given that, I think the sex ratio of 4 has to be at the top end of the plausible range.

The US medical insurance paper had a catchment population of 50 million people, and it analysed medical diagnosis data separately for people given a diagnosis of CFS and of ME (with similar sex ratios found in each). It's a decent paper. The data is analysed in 10 year age bands, and in no age band does the sex ratio (females to males) exceed 2. So, it's possible to discount the data in the age bands above 70 years or even 60 years, and still conclude that the sex ratio in this population was not higher than 2. It's a valid data point, comparable to the English ICD code data in credibility.

 

There is the interchangeability of the ME and CFS diagnoses, and the varying popularities of the two diagnoses over time. It seems likely to me that the higher prevalence of an ME diagnosis aged 80+ could be explained by those people receiving that diagnosis back in the 1980s, when they were in their 30s and 40s. Whereas the diagnosis of CFS became more common in the US (and indeed everywhere) after that time.

The data in the paper was for 2011-2016, and 2017. Note that some people had both CFS and ME diagnoses.

Also note, the paper is not about when a diagnosis was first made, so it's not about teh time of disease onset. It's about people who consulted a doctor and for whom the insurance claim for that consultation had attached an ME or CFS diagnostic code.

I think it might be possible to argue that the people diagnosed with ME in the US after the promotion of the name 'CFS' are actually more likely to have a disease with PEM than the people in those other databases.


More on the prevalence in the older age groups:
It's not inconceivable that ME/CFS is associated with longevity. There's the risk of suicide of course, but, as a group I expect we eat more healthily than most, smoke less, drink less alcohol. We probably mostly don't do risky activities, we drive less. Perhaps that is a small part of the reason why people with a diagnosis of ME made up a bigger proportion of the geriatric population than expected.

It's also not inconceivable that it is true, that the risk of having ME/CFS actually does increase as we get older. I think the illness is quite a bit like being old prematurely. Maybe almost all 90 year olds actually have ME/CFS! We have seen that repeat Covid-19 infections result in an additive risk of developing Long Covid. So, more infections means more rolls of the dice. With waning immune systems, an increasing burden of chronic infections and many older people collected together in retirement homes, perhaps the dice is being rolled more quickly in later years.

But, of course it's true that doctors may have trouble differentiating ME/CFS from old age, and so the prevalence figures in those older age groups can be questioned. However, as I noted, we can set the figures in those older age bands aside when looking for evidence of sex ratios. That data for older age groups does not invalidate the sex ratios found in younger age groups.