Arsenic in Brown and White Rice

I recently added more rice into my diet and wanted to understand the nutritional value of switching from white to brown rice. Then I recalled hearing about the potential risk of more arsenic so I dug into the research.

This 2016 risk assessment from the FDA provided baseline numbers of the amount of arsenic in rice and rice products as well as a quantitative estimate of lung and bladder cancer risk from long-term exposure of these products.

We found that average concentrations of inorganic arsenic – the more toxic form of arsenic – were as follows:

• 92 parts per billion (ppb) in white rice

• 154 ppb in brown rice

• 104 ppb in infants’ dry white-rice cereal

• 119 ppb in infants’ dry brown-rice cereal

The primary culprit of damage is inorganic arsenic in foods, which I learned in not a traditional heavy metal but a "toxic metalloid". The good news is this means inorganic arsenic doesn't bioaccumulate the same way other heavy metals do. The risk however is still continuous exposure over a long period of time which can damage DNA, the skin, and the vascular system.

Dr. Michael Greger also provides a good summary and insights in his video

Which Rice Has Less Arsenic: Black, Brown, Red, White or Wild?Evaluating differences in rice, we can observe:

• brown rice contains more arsenic than white rice

• long grain contains more than shorter grains, though this may be due more to where these grains are grown

• arsenic accumulates in the rice bran, which is why brown rice contain so much more arsenic

• many of the varieties of colored rice were worse than white rice for the reason above

• children under 6 should almost definitely avoid rice AND rice products, with regulations in the EU now adding more conservative limits on the amount of acceptable arsenic in rice products for infants

One bit of good news is brown rice MAY make the arsenic less bio-available as shown in this population study. As Dr. Greger says:

So, if you test the urine of thousands of Americans who don’t eat rice at all, they’re still peeing out about eight micrograms of toxic, carcinogenic arsenic a day. It’s in the air; it’s in the water; there’s a little bit in nearly all foods. But, eat just one food—a cup or more of white rice a day—and your exposure shoots up 65%.

Okay. But, what about those who eat a cup or more a day of brown rice, which technically contains even more arsenic? Your exposure shoots up the same 65%. No difference between the urine arsenic levels of white rice-eaters compared to brown rice-eaters. Now, this was not an interventional study, where they feed people the same amount of rice to see what happened, which would have been ideal. This was a population study. So, maybe the reason the levels are the same is that white rice-eaters eat more rice than brown rice-eaters. So, that’s why they ended up with the same levels? We don’t know, but it should help to put the minds of brown rice-eaters at rest.

So what is the risk? This analysis from the EPA used a Bayesian approach to estimate the dose response of inorganic arsenic exposure from water. It found a linear relationship of more arsenic exposure with bladder and lung cancer. If we extrapolate the EPA's 2025 upper-bound estimate: one daily cup of brown rice for life adds up to 0.42 percentage points of bladder or lung cancer risk (+5.5% relative to a 7.6% baseline); white adds 0.25 (+3%). At three cups: 1.45 (+19%) for brown and 0.8 (+10%) for white.

A more direct causal study would need to prove out this relationship, but this definitely took me aback. For rice staple cultures, three cups of rice daily is pretty normal. These upper bound figures begin to look similar to increased risk of cancer from a daily beer or alcoholic drink.

Unfortunately rinsing rice does little to remove arsenic, especially in brown rice. But researchers did find a parboil method could dramatically reduce the amount of arsenic, removing 54% & 73% iAs from brown & white rice. This is a specific method that requires a lot water that is then tossed out but it does retain a lot of the same nutrients.

My takeaway? We are exposed to arsenic across a number of ways; through our air, water, and food. The increased cancer risks are linear from continued exposure, which means the amount of arsenic consumption scales with risk. If I was a regulatory body in a country where rice is a staple food, I would set significantly stricter regulations to evaluate and limit the amounts in rice and rice products.

Personally, it doesn't seem worth eating rice regularly, especially brown rice. I'll probably try to have less than 3 to 4 servings a week and mix in other grains. I'd be especially careful with infants and rice products that may concentrate the arsenic levels.

Spanking and Physical Punishment Effects on Children

I had a recent discussion with a friend on the effectiveness and/or trauma caused by physical punishments for misbehaviour for children. The most pointed to paper by articles and AI alike is this 2016 meta-analysis titled Spanking and Child Outcomes: Old Controversies and New Meta-Analyses. It provides the best view on corporal punishment effects and effect sizes while differentiating between spanking/light physical punishments and what would be deemed abusive behavior. The paper reviewed two concerns about the existing research:

1. Spanking Has Been Confounded With Harsher Forms of Physical Punishment

2. Spanking Has Only Been Linked With Negative Child Outcomes in Cross-Sectional or Methodologically Weak Studies

By coding the the types of physical punishments, the meta-analysis removed abusive behavior from the analysis and built bivariate links; relationships between spanking and one single outcome. It compared more spanking, less spanking, and in some studies no spanking. By coding papers based on type like cross sectional vs longitudinal, it evaluated outcomes based on the study type. Its conclusion:

The meta-analyses presented here found no evidence that spanking is associated with improved child behavior and rather found spanking to be associated with increased risk of 13 detrimental outcomes. These analyses did not find any support for the contentions that spanking is only associated with detrimental outcomes when it is combined with abusive methods or that spanking is only associated with such outcomes in methodologically weak studies.

The detrimental outcomes are interesting to examine, a litany of issues with the biggest including being a victim of physical abuse, child mental health problems, and negative parent-child relationships.

Two other studies reinforce this finding. A 2021 review in the Lancet stated:

Our review of prospective longitudinal studies has shown that physical punishment is linked with increases in negative child outcomes. Many of these studies used statistical methods to minimise potential confounding and selection bias. The review has documented compelling evidence that physical punishment is harmful to children’s development and wellbeing and has shown no evidence that it is beneficial for children.

And this 2025 meta-analysis found similar results when examining studies focused on lower and middle income families. This controls for confounding factors like norms or socioeconomic factors.

It is not surprise that the read from AI on the matter is that overwhelming strength of evidence lies against physical punishments, often in line with UN calls to uphold children's human rights against this type of means of compliance.

That being said, I retain some skepticism on the results. As the 2016 meta-analysis pointed out:

parents’ use of spanking is not easily or ethically studied through an experimental design, as children cannot be randomly assigned to parents with varying predispositions to spank, nor can parents typically be randomly assigned to spank or not spank...The circumstances of experimentally manipulated spanking thus are likely to be unusual, leading to concern that experiments with parental spanking may suffer from a lack of external validity.

The study says their results show little reduction in negative outcomes when comparing weaker cross sectional studies against more powerfully designed prospective longitudinal studies but I find this reasoning limited because prospective longitudinal studies still lack the ability to control for some confounding factors like a child's baseline behavior. The directionality of outcomes and spanking is in my view still a question and not well controlled.

Coding of physical punishment was also largely self reported. I question the strength of this coding as the ambiguity between abuse and controlled punishment is not verified nor blinded.

Finally, a 2005 study controlled for cultural norms. In countries where physical punishment is more common, it found the negative outcomes dampened but still persistent. With examples like Singapore that still incorporate caning as a punishment means for extreme cases, I think the literature has gaps on how well managed punishment systems effect outcomes.

TLDR; My read

The evidence is clear that the higher end of physical punishment in terms of harshness and volume is strongly associated with negative outcomes for the child. And these observations persist across cultures and socioeconomics. The line between physical punishment and abuse is thin and I believe easily crossed. This is a two way bridge for the evidence, as I think that makes controlling for limited and well defined proportional physical punishment difficult to design and assess.

The 2016 meta-analysis was clear in its conclusion, but the majority of the evidence it examined was showing impacts of more spanking vs less spanking and wasn't designed to analyze spanking vs non-spanking. I think it is clear that at best physical punishments should be used as sparingly as possible if at all. However, I don't think its fair to say the evidence discounts any physical punishment as the ability to create experiments is infeasible. I'd also contend that that conclusion of no physical punishments is culturally loaded and invites societal bias and judgement. As a rule parents should exhaust other means of punishment first.

Microplastics from Keurig K-cups or Capsule machines

I know that a leading factor for microplastic leaching is heat, especially hot water, interacting with plastic. So I dug into what we know about plastic K-cups and microplastics. Effectively there is no research on the topic that I could turn up, which is pretty surprising. The closest study I could find was this 2025 study on beverage products from the UK. This study was linked to by a number of articles like this from microplastics app or Do Coffee Pods Release Microplastics? What the Research Actually Says that cited ~43 microplastic particles (MP) per liter. These article findings fed into AI results like Claude and Gemini, but I found the takeaways provided were incorrect.

The main finding from the study:

Hot coffee prepared in 3 domestic machines (Table SI-3) showed considerable variation in MP levels, with the oldest model (8 years old with water tank made of PET) releasing higher MPs concentrations (17 ± 2 MPs per cup) compared to newer machines (1 and 0.5 years old with ABS water tank releasing 10 ± 2 MPs per cup and 7 ± 2 MPs per cup, respectively). This is consistent with existing research indicating that material degradation over time leads to increased MP shedding (Luo et al., 2022). The higher MP concentrations in coffee prepared using the older coffee machine may be due to wear and tear of the plastic water tank and other materials, a trend also observed in other products like water bottles (Schymanski et al., 2018).

This ended up being one of the largest amounts of microplastics found compared to the other beverages except for Hot Tea. The other comparisons being iced coffee, iced tea, soft drinks, energy drinks, and juices. The problem with this study being repeated for Kcups is a few fold:

• The 43 MP figure was for the Hot Coffee category overall which included other sources like hot coffee from large chains served in takeaway cups, hot coffee made in sachets served in glass cups, and coffee made from capsules.

• The capsules weren't specifically specified as Kcup, this being a UK study, these are more likely nespresso pods. I'll email the paper author to see if I can get the exact types as its not listed in the study and attach what I hear in the comments

• When comparing the capsule results to other Hot Coffee sources, the capsules didn't perform worse overall when you scale the figures above to MP/liter. The authors suggested differences came from degradation of the water tank rather than the capsules themselves. But because we don't know if these were Kcups, we can't be sure this neutral outcome means the pods are the same.

• Some of the articles I listed said the results showed Polypropylene (PP) in the microplastic results which is the primary material for Kcups. However, the authors said this is probably from leaching from the takeaway cup lids in the Hot Coffee category, which is a repeated result from other studies.

• This study only tested for MP sizes greater than 10μm and excludes cellulose, organic microplastic. This makes these results less comparable to prior studies that may have found more microplastics.

TLDR; There's no studies I've found showing Kcup generated microplastics, the amount generated is unknown. Further articles claiming results are likely incorrect or misleading. The biological impacts of microplastics remain debated with unclear risk thresholds for consumption. But I'll leave what the UK study says about impacts below.

Microplastic (MP) ingestion raises health concerns due to their potential to cross biological barriers and accumulate in tissues. Studies suggest that MPs can physically damage gut cells, alter the gut microbiota, and trigger inflammatory responses, which may lead to immune dysregulation and metabolic disturbances (Cox et al., 2019; Deng et al., 2017). MPs can also act as carriers for hazardous chemicals, including heavy metals and endocrine-disrupting compounds such as bisphenol-A, which may leach into the body upon ingestion and disrupt hormonal functions, affecting processes like reproduction and metabolism (Gallo et al., 2020; Hahladakis et al., 2018). The degree of impact is influenced by MP size, shape, and polymer type, with evidence from studies conducted on polystyrene, polyethylene, and polypropylene MPs suggesting that smaller particles and fragment-shaped MPs are more prone to tissue penetration and chemical release upon ingestion (Banaei et al., 2023). The cumulative effect of MPs and associated chemicals in human tissues highlights the importance of establishing standards to limit MP exposure in food and beverages, given the potential for long-term health risks (Schwabl et al., 2019).

X-posting for our Malaysian readers

Health Impacts of Haze from Indonesia Pt. 1

The haze from the Indonesian fires continues its hazardous pass through Malaysia and Singapore, causing alarm and frustration from impacted residents. Walking around Singapore has been a crazy experience as the visible haze feels dystopian compared to Singapore's usual hyper cleanliness. I wanted to deep dive into the known health impacts and what exactly to expect from exposure in this series.

PSI/API Index and PM2.5

Both Singapore and Malaysia use a pollutant index to warn citizens on the risks of air pollution. Singapore's Pollutant Standards Index (PSI) and Malaysia's Air Pollutant Index (API) both calculate a number of air pollutants and translate the impacts to a 0 to 500+ scale. Both denote unhealthy levels starting at 101, very unhealthy above 200, and hazardous above 300. Malaysia declares a Haze Emergency Declaration at 500. Singapore peaked this year at 164 while some regions in Malaysia saw numbers as high as 476 to 520 such as Serian, Sarawak.

While the two systems have slightly different ways of calculating the index, both measure PM2.5 among others air pollutants. PM2.5 calculates the amount of fine atmospheric particulate matter at or below 2.5 micrometers in the air. This is the most relevant group of pollutants to track as it measures the chemicals related to the haze stemming from Indonesia's smouldering peatland fires. PM10 similarly measures particulate matter, but larger at or below 10 micrometers. PM2.5 is more toxic and harmful due to its smaller size, making them more likely to be absorbed in the bloodstream and more chemically active due to the greater surface area.

This 2017 study found the chemical makeup from haze is dominantly Organic Carbon (OC) with Polycyclic Aromatic Hydrocarbons (PAHs) and toxic heavy metals like Cadmium, Nickel, Lead, and Chromium. The latter two categories make up a much smaller portion of the haze and while repeated exposure is not good, it is less of a concern than OC which dominates the mix.

What may be more alarming is this haze is categorized as aged smoke vs raw smoke, chemically altered by UV exposure from the days the smoke took to travel to the impacted areas. As this paper examines this could lead the chemicals in the haze to cause more oxidative stress on the system, though not all studies align to this conclusion.

Known Health Impacts - Respiratory and Cardiovascular

I will deep dive into this a bit more in a subsequent post, but the tldr is there are limited long-term health studies that can give us good evidence on repeated exposure and cancers or cardiovascular disease.

What is clear are the short term impacts. Haze exposure even in the moderate zone can lead to respiratory infections and illnesses. In Malaysia, where the haze is more severe, higher unhealthy zones can lead to respiratory fatalities. This study in the Klang Valley found that haze days were associated with a 19% rise in natural-cause respiratory mortality overall, and higher among specific groups: roughly 30% for children under 14 and about 30% for men on the day of exposure. Effects persisted a day or two after.

Singapore is not safe from haze related fatalities though. While haze effects are less and of shorter duration, moderate to unhealthy levels can still lead to greater cardiovascular issues like heart attack and stroke. That's why Singapore warns citizens to avoid rigorous exercise during unhealthy haze periods, especially if you have underlying pre-existing issues.

X-posting my KCup microplastics research here

[Project Credo Monthly Broadcast] July/August 2026

If you're new here, say hello and reach out to the team at [email protected]. We've been hard at work improving all aspects of the site so we put our best foot forward for anyone new to the site.

Community

A few new boards have been created to help us fill out the types of conversations we want to read about.

• Podcasts We Love | Project Credo covers a ton of interesting factoids and discussions across all the podcasts we've been listening to

• Impact of Social Media on Society | Project Credo will be a vital place we have conversations about how social media has been reshaping our lives, important for us to catalog research to keep our users safe and happy

• Influencer Science-Check | Project Credo gives us a community to do one of the biggest asks we get, fact check all the crazy things we hear on the internet!

• Parenting | Project Credo is a space we know people have a lot of questions, somewhere we want to bolster more information for parents looking for good science based content

Features and Bug Fixes

• Reference summaries before you post! Now before you hit post, you can read the summaries and metadata for the papers and links you add.

• Model benchmarks. To help ensure we're providing the best quality across Project Credo, we're running benchmarking across different models to test the accuracy of our summary feature. More of this to come for all the AI powered features on the site.

What we're working on

• Board agents: We want to enable more automation to make boards even more valuable. We're testing soon!

• Live Chat: We've heard across different groups that they mostly get their research information from peer chat groups. We're looking into enabling this functionality to bring over more power users.

Shingles Vaccine Protects Against Dementia

A flagship study came out in Nature in 2025 that solidified a long observed finding, that the Shingles (Herpes Zoster) vaccine can protect against dementia.

Previous studies were classic observational cohort designs which can fail to show causation or account for confounding factors. This 2021 study found "a combined analysis estimated a 31% lower risk" and indicated the need for stronger research.

In comes the 2025 Nature study, this time a natural experiment arose:

To provide causal as opposed to correlational evidence, we take advantage of the fact that, in Wales, eligibility for the zoster vaccine was determined on the basis of an individual’s exact date of birth. Those born before 2 September 1933 were ineligible and remained ineligible for life, whereas those born on or after 2 September 1933 were eligible for at least 1 year to receive the vaccine.

The finding:

We show that receiving the zoster vaccine reduced the probability of a new dementia diagnosis over a follow-up period of 7 years by 3.5 percentage points (95% confidence interval (CI) = 0.6–7.1, P = 0.019), corresponding to a 20.0% (95% CI = 6.5–33.4) relative reduction. This protective effect was stronger among women than men.

Interestingly, the newer recombinant vaccines seem to be ever more effective than the live-attenuated vaccines (the ones tested in the Nature experiment), though more studies are in the pipeline to deepen this finding.

The mechanism for why the protective effect happens is still TBD, though some hypothesize that detrimental impact of getting Shingles on the brain may cause long term damage that leads to dementia.

Either way, the impacts seem clear, the Shingles vaccine seems like a must-get in a full range defense against dementia.

Social Media and Child Safety

It was recently reported that Meta was ordered to pay $567M in additional fines for an abatement program in New Mexico due to its failure to warn the public about its platforms' risks to children. The fines were in addition to $375M ordered against Meta for a similar earlier claim in New Mexico. This marks the largest child safety ruling against Meta and will have reverberations for social media companies in the future.

The court case ruled Meta acted as a public nuisance infringing on the health and safety of the public "knowingly" and "without lawful authority". The state built its case around these 10 components:

1. New Mexico is experiencing a youth mental health crisis.

2. Problematic social media use is a cause of the youth mental health crisis.

3. Meta’s platforms are widely used in New Mexico.

4. Meta implemented platform features that were designed to maximize engagement and time spent on its platforms, for all users and for teenage users in particular.

5. Meta’s engagement features encourage and facilitate problematic interactions between adults and children.

6. Harm caused by Meta’s products affects a significant number of people in New Mexico.

7. Although many social media platforms compete for the attention of New Mexico’s youth, Meta’s platforms rely on methods for optimizing engagement that disproportionately impact education and interpersonal relationships

8. Meta and its platforms provide a valuable service in New Mexico despite these harms, and it is not the only social media company impacting youth.

9. Meta knowingly designed its platforms to maximize engagement, including in ways that are harmful to teenagers, and does not adequately disclose the risks of harm to its users.

10. Meta’s conduct is a cause of and significantly contributes to the youth mental health crisis in New Mexico.

Deep dive into evidence of harm towards youth

The court case itself didn't provide any specific papers as far as I can tell, only references to HHS advisories, reports, and internal Meta research. The evidence was built on expert testimony, which the court found strong enough to state:

Even without the evidence and testimony that certain percentages of youth ... would not have those conditions but for social media use, the highly probable result of Meta's engagement optimization practices ... is that such practices would increase anxiety, depression, and sleep deprivation among teenagers.

The experts on both sides provided counter testimony on their view of the connection between social media and harm to youth. Dr. Jean Twenge testified that

most longitudinal studies confirm that adolescents’ social media use leads to depression ... and that experimental studies show that people who eliminate or reduce social media are less lonely and depressed, which further confirms the causal path between social media and teen mental health harms.

While the evidence wasn't listed, a 2022 study Social Media and Mental Health backs this testimony showing that the arrival of Facebook on college campuses lead to worse student mental health. The HHS advisory on Social Media and Youth Mental Health also listed this 2019 JAMA Psychiatry study.

Meta on the other hand pointed to a 2024 consensus report on Social Media and Adolescent Health which pointed to other factors causing overall mental health decline and a complex role social media plays in society.

My take

There does seem to a slew of evidence pointing to correlation between social media and declining mental health with teens and adolescents. Additionally, the ability for online predators to connect to underage minors is also blatant. Arturo Bejar, who worked as Meta’s Senior Director of Engineering in charge of the Protect and Care team stated:

the product is very good at connecting people with interests, and if your interest is little girls, it will be really good at connecting you with little girls.

But I agree with the primary conclusion from the Social Media and Adolescent Health report. There are benefits that are associated with social media including connection to friends and combating isolation, civic engagement, learning, and self expression.
The issue arises when we try to simplify the debate to net benefit or harm. Instead, its better to consider the piecemeal impacts and combat clear harms while retaining clear benefits. I don't think its helpful to see social media as a pure harm, just the same I wouldn't categorize screens and screentime as a pure harm.

[Project Credo Monthly Broadcast] June 2026

If you're new here, say hello and reach out to the team at [email protected]. We spent the last month reviewing how to stretch our organic reach across different platforms to integrate more users onto the site. We saw mild pickup of new users, but we're definitely struggling to pull in new consistent users.

Community

We are trying a few new board styles and trying to see how well smaller posts do. This is so new users feel less activation energy to create their first post. We are also reviewing new post styles that can be both engaging but also smaller action cost for users.

Features and Bug Fixes

• Paper detection: We've built identified more fluid ways to identify paper links so we can automate URL to paper lookups where possible.

• Improved references: We've just pushed a new layout for references to make absorbing the high density information more easily.

What we're working on

• Board bots: We are testing the ability for board owners and contributors to utilize board bots to find papers and auto generate summaries giving board owners super powers in better paper understanding.

• More layout improvements: We want to streamline our UI/UX a bit more to make all interactions easier.