Vitamin and mineral supplements and fatigue: a prospective study 2025 Xie et al

Abstract

Purpose
The consumption of vitamin/mineral supplements (VMS) and vitamin/mineral and/or dietary supplements (VMDS) is popular among the general population. However, the association of VMS/VMDS with fatigue remains sparse and conclusions are mixed. We aimed to understand the association between VMS/VMDS and fatigue.

Methods
Prospective study in the city of Lausanne, Switzerland, including 1361 participants (50.3% female, mean age 61.0 ± 9.4 years). Participants were divided into VMS/VMDS users and non-users. Fatigue levels were assessed using the Fatigue Severity Scale (FSS) and the 14-item version of the Chalder Fatigue Scale (CFS). Statistical analyses included multivariable logistic regression for categorical outcomes and analysis of variance for continuous outcomes, adjusting for relevant covariates.

Results
No association was found between VMS consumption and changes in FSS (mean ± standard error 0.05 ± 0.03 vs. -0.06 ± 0.14 for non-consumers and consumers, respectively, p = 0.440) and CFS (-0.05 ± 0.06 vs. 0.22 ± 0.28, p = 0.388). Similarly, no effect of VMS consumption was found on incidence odds ratio and 95% confidence interval: 1.75 (0.82–3.74), p = 0.149 or remission 1.36 (0.49–3.74), p = 0.550 of clinical fatigue. Similar findings were obtained for VMDS: FSS 0.06 ± 0.04 vs. 0 ± 0.08, p = 0.577; CFS − 0.07 ± 0.08 vs. 0.04 ± 0.15, p = 0.545 for non-consumers and consumers, respectively. OR 1.96 (1.20–3.20), p = 0.008 and 1.14 (0.57–2.31), p = 0.712 for incidence and remission of fatigue. Alternate or persistent VMS/ VMDS consumers had a higher incidence of clinical fatigue and a higher increase in FSS compared with never consumers.

Conclusion
In this population-based sample, we found no consistent association between VMS or VMDS consumption and remission of fatigue. Conversely, VMDS users tended to develop greater fatigue.

Open access

 

Or the people that develop fatigue are more prone to experiment with supplements to try and deal with the fatigue.

 

Am I misunderstanding something? Because this study design doesn't seem useful for figuring out if vitamins help or don't help fatigue. As far as I can tell, they didn't ask about new vitamin usage. Many of these people may have been taking vitamins for long before this study started, and if there was benefit for their fatigue, presumably it may have already stabilized. I'm not sure why they'd expect to find significant effects even if vitamins were beneficial.

It doesn't seem different from if they asked a bunch of people about whether they use statins and check cholesterol levels five years apart, saw that they didn't decrease, and claimed that statins don't reduce cholesterol. Many of these people will already have stabilized at a lower cholesterol level from statin use before the study, so further decreases wouldn't be expected in their followup timeframe.

And further, it seems like it would be very unlikely to see a significant effect when lumping every single vitamin and mineral together. Not really different from, for example, asking people about whether they take any medication at all and checking their blood pressure, then saying medication doesn't seem to provide any benefit for high blood pressure, even though they lump pain killers, insulin, antivirals, immunosuppressants, and antihypertensives all together into "medications".

Why is that unlikely?

 

@forestglip I don’t have any answers for your questions. It seems like a very flawed and biased paper.

 

I agree. Untreated I have a B12 deficiency, and if in deficit B12 reduces fatigue, but when regularly taking B12 I am not in deficit so although doses may stop me going into deficit they have no impact on fatigue.

We need to know what the vitamin levels an individual already has, what doses are being taken and how this relates to an individual’s requirements. I suspect vitamin supplements are only of use if needed to prevent deficit, but this study does not answer even that, let alone any questions about any postulated benefits arising from dosages at higher levels.

[edited to add final sentence]

 

What's funny is how "supplement stacks" were a 100x bigger thing for the first two years of Long Covid than even the most hyped "brain retraining" stuff out there. People were sharing long lists of supplements, their dosage, posology, and so on. Before any other type of treatments were ever discussed, this was "it" in the LC communities. This was how you recovered.

Then it died down completely. Almost no one speaks of those anymore. What used to be the most common thing to share is now very rare. Because they don't work, and since this is a community that needs things that work, it just moved on. And yet it would be easy to get results that are just as 'promising' as any of the psychobehavioral crap. All you need is the right amount of bias. And with the highest level of bias, such as we see with the most kooky 'brain retraining' stuff, it would likely rate even better. Slightly, barely better than nothing, but it would rate. It's easy to get slightly better than the smallest blip of a signal.

Because determining what works and what doesn't is very hard, and it's very easy to apply just the right amount of bias to get the answers you want. Here we see an example where the bias isn't in favor of it, and it finds no benefits. But none of the psychobehavioral crap is any better, and yet it's been subject to decades of assertions that it works very effectively, and this hasn't stopped despite this entire ideology having been exposed as a fraud. Because the 'research' backing it up is basically three biases and a conflict of interest in a trenchcoat.

And ironically there probably are a few supplements that do help some. Some, who happen to have a bit of deficiency. The whole idea behind doing analyses for 'responders vs non-responders', none of which applies to the woowoo stuff. But this won't have decades of obsessive tweaking and tuning, so it will stop here. Because of the bias and how it aligns.

It's all bias. Without science, you literally only get biases, and scientists are just as biased as the rest of the general population. They have fewer of them, but what they do have makes up for count by being applied to an extreme level, backed by intransigence and hubristic dogma of the very same type that tends to make up conspiracy communities: having a vastly inflated sense of self-importance and belief in the superiority of their, well, beliefs.