Is fish oil harmful to your heart?
Is fish oil harmful to your heart?
More concerns are being raised that fish oil may increase the risk of atrial fibrillation and other heart-related issues, but is there merit to this worry?
Edit 6.1.2024: Grammatical changes were made in one sentence below. The sentence that was changed now states the following:
For one, the researchers didn’t take into account how much fish oil people were consuming- everyone considered “regular users” of fish oil were based on some subjective measure made by the researchers (which they don’t cite).
In another case of the media not scrutinizing the studies they cover here’s another one to add to the list.
This time, a recent study seems to suggest that supplementation with fish oil may increase the risk of certain cardiovascular complications. In particular, atrial fibrillation as well as the possible increase in stroke risk are the two main concerns highlighted from this new study, leading the media to brew up another hysteria-filled storm:
Now, before you start tossing out all of your fish oil let’s take a look at why some of these remarks may be unfounded and plagued with issues.
Before we begin let’s consider the ties between fish oil use and cardiovascular risk. It may surprise readers, but fish oil has come under scrutiny in recent years as some evidence has come out suggesting that fish oil use may be associated with atrial fibrillation- a disease in which abnormal electrical impulses within the atria of the heart causes irregular pulse rates.
Although there is growing research suggesting an association between fish oil and atrial fibrillation the overall evidence is rather mixed.
Some bits of evidence for further reading can be seen below:
Gencer, et al.1- Effect of long-term marine Omega-3 fatty acids supplementation on the risk of atrial fibrillation in randomized controlled trials of cardiovascular outcomes: a systematic review and meta-analysis
Lombardi, et al.2- Omega-3 fatty acids supplementation and risk of atrial fibrillation: an updated meta-analysis of randomized controlled trials
Huh, J. H., & Jo, S. H.3- Omega-3 fatty acids and atrial fibrillation
There is more information out there, although the first two citations above are meta-analyses which should lead readers onto individual studies to look out for.
Overall, variability in results seem to relate to differences in doses of Omega-3’s, differences in formulation, differences in subject groups, study methodology, i.e. issues typical of these sorts of studies.
For our sake it’s the study methodology which is of the question for this recently published study. It’s a rather short study to read through so I encourage readers to look at the paper for themselves while I highlight a few of the noticeable pitfalls.
Data was based on over 400,000 people enrolled in the UK’s Biobank study. Participants were between the ages of 40-69 years and had no prior cardiovascular events.
Participants were categorized based on baseline questionnaires in which they described their fish oil usage:
Information on regular use of fish oil supplements was collected from a self-reported touchscreen questionnaire during the baseline survey.14 15 Each participant was asked whether they regularly used any fish oil supplement. Trained staff conducted a verbal interview with participants, asking if they were currently receiving treatments or taking any medicines, including omega 3 or fish oil supplements. Based on this information, we classified participants as regular users of fish oil supplements and non-users.
Participants were followed as part of the Biobank study until death, loss to follow-up, or the end of the study period. Health records and death certificates of participants were examined in order to examine incident rate of the various cardiovascular issues.
Participants were followed up from the time of recruitment to death, loss to follow-up, or the end date of follow-up (31 March 2021), whichever came first. Incident cases of interest, including atrial fibrillation, heart failure, stroke, and myocardial infarction, were identified by linkage to death registries, primary care records, and hospital inpatient records.11
There are already a few problems with this methodology. For one, the researchers didn’t take into account how much fish oil people were consuming- everyone considered “regular users” of fish oil were based on some subjective measure made by the researchers (which they don’t cite). Also, as is typical of many survey-based, long-term studies the researchers didn’t seem to follow people’s use of fish oil throughout the study period.
It's highly unlikely that participants will use the same dose of fish oil, same brand, or will continue to use/not use fish oil during a period of over 11 years.4 This seems to be a rather common issue in these types of studies where there doesn’t appear to be clear distinctions between the two groups. Again, for a study that encapsulates cardiovascular events over the course of 11 years can one really assume that people within either group are constantly static in their fish oil use/nonuse? What exactly is the fish oil status of individuals at the time of their cardiovascular-related incident?
And remember that this doesn’t take into account the fact that other circumstances could influence health outcomes, such as the fact that some people in both groups may become hypertensive, may become obese, along with other possibilities.
Remember that there may be reasons for people to take fish oil as well, as some people may take it for its possible health benefits but otherwise be healthy while some people may be more prone to cardiovascular disease and therefore may want to find ways of reducing their risk. Why someone may choose to take fish oil may also explain differences in outcomes.
If researchers are unsure, then you can’t exactly tie these two variables together.
It’s because of these issues that the results should be under question.
Researchers broke down participant events based on various transition patterns. This layout may be a bit confusing but note that it shows movement from baseline “healthy” status to other areas depending on the type of event that occurred. For instance, “Transition A” is a move from “healthy” to an incident of “atrial fibrillation”. The main difference between the transition patterns is the event listed on the far-right, with the researchers swapping MACE for other events such as strokes, heart failures, and myocardial infarctions. So when looking at Table 2 note that the differences between the transition pattern groups lie mostly in different outcomes of “Transition B”, “Transition D”, and “Transition F”.
A representative for transition patterns I is shown below:
When comparing incidences in those taking fish oil relative to non-fish oil users the authors noted a slightly higher incidence of atrial fibrillation within the fish oil group, as well as very minor increased incidence of stroke with a barely passable p score:
Remember that these values are based on flawed data collection, and even then this is not quite an egregious increase.
There are also a few issues regarding the numbers provided here. Note that all 4 transition patterns note similar numbers, hazard ratios, and p values for “Transition A” from baseline→ atrial fibrillation.
However, the values for “Transition C” and “Transition E” are inconsistently listed across the 4 transition patterns even though these values should be the same. There’s no mention why these values are different between all of the groups so it’s curious why these differences are occurring. This isn’t necessarily related to the issues with the hazard ratios, but at least raises questions regarding how this data is collected and pieced together.
Overall, we can see why there are several issues with this study and why some skepticism should be taken when looking at these results.
As of now the link between fish oil and atrial fibrillation is unclear. More importantly, if research looking into this link rely on flawed studies such as this one then we should be more inclined to show hesitancy in any data being put forth.
When looking through the data bear in mind differences in study methodology and how this may influence outcomes. It’s likely that a combination of highly varied and poorly designed studies may explain why the link is still unclear.
So rather than take the word of journalists who may not read the studies they cover always remember that you are capable of doing your own research and come to your own conclusions.
If you enjoyed this post and other works please consider supporting me through a paid Substack subscription or through my Ko-fi. Any bit helps, and it encourages independent creators and journalists such as myself to provide work outside of the mainstream narrative.
Gencer, B., Djousse, L., Al-Ramady, O. T., Cook, N. R., Manson, J. E., & Albert, C. M. (2021). Effect of Long-Term Marine ɷ-3 Fatty Acids Supplementation on the Risk of Atrial Fibrillation in Randomized Controlled Trials of Cardiovascular Outcomes: A Systematic Review and Meta-Analysis. Circulation, 144(25), 1981–1990. https://doi.org/10.1161/CIRCULATIONAHA.121.055654
Lombardi, M., Carbone, S., Del Buono, M. G., Chiabrando, J. G., Vescovo, G. M., Camilli, M., Montone, R. A., Vergallo, R., Abbate, A., Biondi-Zoccai, G., Dixon, D. L., & Crea, F. (2021). Omega-3 fatty acids supplementation and risk of atrial fibrillation: an updated meta-analysis of randomized controlled trials. European heart journal. Cardiovascular pharmacotherapy, 7(4), e69–e70. https://doi.org/10.1093/ehjcvp/pvab008
Huh, J. H., & Jo, S. H. (2023). Omega-3 fatty acids and atrial fibrillation. The Korean journal of internal medicine, 38(3), 282–289. https://doi.org/10.3904/kjim.2022.266
The median follow-up period was 11.9 years, so there’s even a huge degree of variability in follow-up. Remember that follow-up doesn’t mean the time from study inclusion to first incident i.e. first instance of atrial fibrillation. Instead, follow-up likely means that when a death is recorded or when the study ended researchers examined participant data to check for records of reported cardiovascular incidences. So, over the course of nearly 12 years multiple events may have occurred.
Thank you. My initial impression was that all recent press about "hamrful" fish oil was unproven nonsense and this article confirms it for me.