Summary

• I calculated D3, long-chain omega-3, and multivitamin-multimineral supplementation are worth it for me for a future annual net income above 30.2 k, 279 k, and 121 k$. I did not cover benefits from increased future annual net income.
• Taking cheap supplements of debatable benefits may be worth it sometimes, but not always. I would like to see a greater focus on cost-benefit analyses to determine the break-even conditions.

Methods

Overview

Here are my calculations.

I calculated the future annual net income above which supplementation is worth it for me from the ratio between:

• The annual cost of supplementing.
• The annual life extension caused by the supplementation. I obtained this multiplying:
  • The annual reduction in mortality caused by the supplementation.
  • My life expectancy.

I did not cover benefits from increased future annual net income.

Annual reduction in mortality

D3

I stipulated 2 kIU of D3 supplementation every day reduces mortality by 5 %. Zhang et al. (2019) got a risk ratio for vitamin D3 supplementation of “0.95 (0.91 to 1.00 [95 % confidence interval])”, and for vitamin D supplementation among people with mean 25 hydroxyvitamin D lower than 50 nmol/L, which is a common threshold for defining vitamin D deficiency, of “0.95 (0.90 to 1.01)”. García-Maldonado et al. (2024) estimated around 60 % of vegans “in the Madrid region of Spain”, which is roughly at the same latitude as Lisbon (where I live), have vitamin D3 deficiency as defined by that threshold (see Figure 2). I assume the fraction would be higher among vegans who do not take a D3 supplement, as I think this is the most recommended to vegans apart from B12, and therefore expect many of the analysed participants to be taking it. As a result, I believe the estimate of Zhang et al. (2019) for people with vitamin D3 deficiency is the most informative to assess D3 supplementation among vegans like me.

The annual risk of dying for my age range in Portugal in 2022 was 0.04 %. Multiplying this by the above relative reduction in mortality leads to an annual reduction in mortality of 20.0 micromorts. 1 micromort corresponds to a risk of death of 1 in 1 M. Travelling 370 km by car respects 1 micromort, under some conditions I did not check. So the annual reduction in mortality amounts to driving 7.40 k km less by car.

Long-chain omega-3

I assumed 1.58 g/d of long-chain omega-3 supplementation reduces mortality by 3.00 %. From Abdelhamid et al. (2020), “Meta‐analysis and sensitivity analyses suggested little or no effect of increasing LCn3 on all‐cause mortality (risk ratio (RR) 0.97, 95% confidence interval (CI) 0.93 to 1.01; 143,693 participants; 11,297 deaths in 45 RCTs [randomised controlled trials]; high‐certainty evidence)”. Abdelhamid et al. (2020) was published on the Cochrane Database of Systematic reviews, and described by the authors as “the most extensive systematic assessment of effects of omega‐3 fats on cardiovascular health to date”. I got the reference dose of long-chain omega-3 of 1.58 g/d from the geometric mean between the lower and upper bound of the range covered by Abdelhamid et al. (2020). “LCn3 [“long‐chain omega‐3”, “including eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)”] doses ranged from 0.5 g a day to more than 5 g a day (19 RCTs gave at least 3 g LCn3 daily)”.

I speculated the relative reduction in mortality is proportional to the logarithm of 1 plus the daily dose of long-chain omega-3 in g. So I got a reduction in mortality of 0.830 % for the 0.3 g/d of long-chain omega-3 in the supplement I have been taking.

Based on my annual risk of death, I estimated an annual reduction in mortality of 3.32 micromorts. This amounts to driving 1.23 k km less by car.

I do not know whether the relative reduction in mortality would be smaller or larger for people like me following a plant-based diet. I searched for studies that could inform this using Elicit, but did not quickly find any RCTs looking into people following a plant-based diet. Abdelhamid et al. (2020) analysed the general population, where there is greater intake of long-chain omega-3 outside supplements, so the effect could be larger for people following a plant-based diet. On the other hand, people following a plant-based diet tend to have a better cardiovascular health, this is thought to be one of the main benefits of long-chain omega-3 supplementation, and I expect improving the health of healthier people to be harder, so the effect could be smaller for people following a plant-based diet.

I have little idea about whether other substances in long-chain omega-3 supplements increase or decrease the benefits. Contaminants would be harmful, whereas other components would tendentially be beneficial. The overall effect of the contaminants and other components will not necessarily be positive. Abdelhamid et al. (2020) found that increasing the intake of alpha-lipoic acid (ALA), which is a short-chain omega-3 fatty acid, increased mortality by 1 %, although there is large uncertainty (“RR 1.01, 95% CI 0.84 to 1.20; 19,327 participants; 459 deaths in 5 RCTs, moderate‐certainty evidence”).

Multivitamin-multimineral

I considered taking 1 capsule of 0.82 g of my multivitamin-multimineral every day decreases mortality by 2.00 %. From Macpherson et al. (2013), “Across all studies, no effect of multivitamin-multimineral treatment on all-cause mortality (RR: 0.98; 95% CI: 0.94, 1.02) was observed”. “The average age of the pooled sample was 62 y, and the average duration of supplementation was 43 mo”. Macpherson et al. (2013) reports “no effect”, but they found a small beneficial effect, as their meta-analytic risk ratio is below 1. There is significant uncertainty about whether the effect is beneficial or harmful, but it follows a continuous distribution, so it cannot be exactly neutral, although it practically can.

Based on my annual risk of death, I estimated an annual reduction in mortality of 8.00 micromorts. This amounts to driving 2.96 k km less by car.

It is again unclear to me whether the relative reduction in mortality would be smaller or larger for people following a plant-based diet. I searched for studies that could inform this using Elicit, but did not quickly find any RCTs looking into people following a plant-based diet.

Life expectancy

My (period) life expectancy is 55.9 years based on mortality rates in Portugal in 2023. However, my (cohort) life expectancy is longer because mortality rates will be lower in the future. Males born in the United Kingdom (UK) in 1997 (the year I was born) are predicted to have a life expectancy at birth of 84 years, which is 9.40 years longer than the life expectancy at birth of males in the UK in 1997 of 74.6 years. I added that difference to my life expectancy for recent mortality rates to arrive at a life expectancy for future mortality rates of 65.3 years.

I do not think transformative AI (TAI) would extend life expectancy much more than predicted by past trends. There is little correlation between real gross domestic product (real GDP) per capita and period life expectancy at birth across countries where this is at least 80 years.

Annual cost

D3

The supplement of D3 I have been taking costed 0.0262 $/kIU on 24 March 2025. For my daily dose of 2 kIU, the daily cost of the supplement is 0.0525 $.

I estimated taking the supplement costs me 10 s each day. I took 40 s to take the last 4 of 5 supplements after my lunch on 3 March 2025^[1]. I am interested in the time saved by taking one less supplement, so the longer time required to take the 1st does not matter. I assumed losing 1 h has a cost of 20 $. So I ended up with a cost of spending time taking the supplement of 0.0556 $/d.

The above imply a daily cost of 0.108 $, which results in an annual cost of 39.5 $. 48.6 % comes from the supplement, and 51.4 % from the time taking it.

Long-chain omega-3

The supplement of long-chain omega-3 I have been taking costed 0.110 $/g on 24 March 2025. For my daily dose of 1 g, which corresponds to 1 capsule, and includes 0.300 g of long-chain omega-3 in DHA and EPA, the daily cost of the supplement is also 0.110 $.

The above plus the cost of spending time taking the supplement imply a daily cost of 0.165 $, which results in an annual cost of 60.4 $. 66.4 % comes from the supplement, and 33.6 % from the time taking it.

Multivitamin-multimineral

The multivitamin-multimineral I have been taking costed 0.143 $/g on 3 March 2025. For my daily dose of 0.82 g, which corresponds to 1 capsule, the daily cost of the supplement is 0.117 $.

The above plus the cost of spending time taking the supplement imply a daily cost of 0.173 $, which results in an annual cost of 63.1 $. 67.8 % comes from the supplement, and 32.2 % from the time taking it.

Results

Discussion

Taking cheap supplements of debatable benefits may be worth it sometimes, but not always. I would like to see a greater focus on cost-benefit analyses to determine the break-even conditions.

I underestimated the break-even future annual donations due to neglecting the greater cost of losing time for greater donations. For my estimates of the break-even future annual donations, losing 1 h would cost more than my assumed value of 20 $.

I am planning to stop my long-chain omega-3 supplementation. I believe I will donate much less than 279 k$/year in the future to my preferred organisations. I guess non-mortality benefits would not sufficiently lower the break-even donations for supplementation to be worth it. Witte et al. (2013), which is discussed in a video from NutritionFacts.org, ran a randomised, double-bind, placebo-controlled trial involving 65 healthy people which found “positive effects on brain functions”, but they do not report any effect sizes based on differences between means^[2]. So it is difficult for me to assess the strength of the effect. I would like to see a preregistered RCT like Sandkühler et al. (2023) estimating the effect of long-chain omega-3 supplementation on the intelligence quotient (IQ).

I intend to continue taking a multivitamin-multimineral. I expect donating less than the respective break-even donations of 121 k$/year, but I want to ensure adequate iron intake while potentially having some added benefits. Having less than 30 ng/mL of ferritin is “low‐certainty evidence” for “iron deficiency in people presenting for medical care”. I had 33.0 ng/ml in my last blood test, and I was already taking a multivitamin-multimineral with iron then. According to the National Health Service (NHS), men aged 19 and over should consume 8.7 mg of iron per day. 1 capsule of the multivitamin-multimineral I have been taking contains 7.3 mg of iron, thus covering 83.9 % of my daily needs.

It is unclear to me whether I will donate more or less than 30.2 k$/year. Yet, I also plan to continue my D3 supplementation, as I worry the non-mortality benefits, which would decrease the break-even donations, may be significant.

1. ^{^}

    Creatine, D3, iodine, long-chain omega-3, and multivitamin-multimineral. I took B12 before breakfast.

2. ^{^}

    They found supplementation “enhanced executive functions by 26%, whereas performance remained constant after placebo (paired t-test, t(31) = 3, P = 0.005; Fig. 3A)”. Based on Fig. 3A, they are saying the mean of the following composite score for executive function among participants in the treatment group increased by 0.26. “[z phonemic fluency + z semantic fluency - z TMT [trail making test] (part B - part A)/partA - z STROOP [Stroop Color-Word test] (part 3 - (part 1 + part 2))/2]/4”, where z refers to the z-score of a person on a test. “Regarding the composite score for memory, both groups showed a similar retest effect at follow-up, with no significant effect of group (ANOVARM, group X time: P = 0.6; time: F1,63 = 19.8, P < 0.001)”. “For sensorimotor speed, a global retest effect was noticed that did not differ between groups (composite score, ANOVARM, time: F1,63 = 14, P < 0.001)”.