Just like the boy who cried wolf- headline writers need your attention and clicks! The villagers in that story first ran to save the boy - which was a waste of energy since there was no wolf, and then angry at the false alarms, decided to not respond when the real wolf came. When it comes to health news - something that concerns us all - instead of either believing everything we read, or not believing anything we read, how cool would it be to become an intelligent reader able to discern between click-bait or fake news, and the real deal. Thus, the goal is to develop our own "wolf monitoring system", as it were.
In this issue, we step through scientist and celebrity generated headlines since those two can catch our attention quite reliably. We end with some tips that we have found useful. Once equipped, it will be easier for us to go deeper into nutrition, especially the forthcoming protein chronicles. In Sabzbag recipe comics, we bring you - Mirza Ghasemi- a delicious eggplant dish from the North of Iran and in the community section, we discuss how the drama or the noise of each month can hide the useful long term eating pattern- the signal - that is quite obvious and responsible for most of our outcomes.
Enjoy! |
|
|
|
|
Figure 1: Some Sample Headlines The Guardian, and Fox News,The red highlighting for the Guardian article rendered by us for clarity. |
A lot of us eat meat, do not want to get cancer, like tomatoes OK, and are at least aware of who Tom Brady and Gisele Bundchen are. Hence we find ourselves clicking on the headlines in Figure 1. Let's begin our fact finding journey with academics and then move on to the celebrities.
Scientists are super smart, passionate and well trained HUMANS
Scientists tend to be incredibly passionate, well trained and hardworking but as Nature magazine says - "Scientists are human." Thus like the rest of us, they have a vested interest in promoting their own work. Luckily, there are other scientists in the community who relentlessly test the claims made by one group. In fact this constant and systematic testing based on the notion of falsifiability IS the heart of science, as Karl Popper - a very famous philosopher of Science - argued.
Why a consensus of scientists or meta-studies?
Now we will step through why, as thoughtful consumers of information, instead of relying on one particular study by one particular scientist, we should look for a consensus - in the form of meta-studies to emerge. When many smart people have replicated the results, or found something similar in different populations, the finding has been tested extensively and is thus a lot more believable. That is when we should start paying attention.
Example I. Academic Studies where almost every food is linked to cancer.
Two scientists picked up a popular cookbook, flipped pages at random to find recipes and selected the most common 50 ingredients used. They found that 40 of those 50 ingredients had been investigated for links to cancer with a total of 264 single studies and 36 combined or "meta-analysis." These ingredients were: veal, salt, pepper spice, flour, onion, celery, carrot, bacon, potato etc. The studies were from February 1976 to December 2011, with the majority (71%) of the articles being published after 2005.
😱 Potential Confusion: Coffee is positively linked to cancer or not?
Figure 1 reproduced from the study shows various ingredients at the bottom. Let's take the case of coffee circled in yellow with each dot representing one study that investigates it for a link with cancer and the vertical line through the center indicates no effect. Each dot on the left indicates one study that finds coffee consumption associated with a lower risk of cancer and each dot on the right finds one study associated with an increasing risk of cancer. About half the dots are on the left (decreased risk) and about half are on the right (increased risk). It's a draw! The effect sizes also vary with some studies finding 300% risk vs. normal and some finding only 30% of the risk of normal. You may see newspaper headlines (NYTimes: "coffee may reduce cancer" and coffee drinkers need cancer warning, judge rules...) associated with this. However thoughtful meta-studies or popular expert blogs like WebMD would say do not change your habit too much. |
|
|
|
|
Figure 2: Schoenfeld JD, Ioannidis JP. Is everything we eat associated with cancer?. The colored circles added by us for ease of exposition. The top half shows the types of cancers investigated and each dot is one study that finds a food ingredient positively or negatively related to those cancer. Thus, breast and gastrointestinal cancers are much more studied than lung cancer for associations with diet. The bottom half shows some specific foods and each dot represents one study that investigated the link between the food and cancer with dots on the left indicating a lower risk and dots on the right indicating a higher (relative) risk. |
|
| |
Figure 3: Shows how combined studies or "meta-analysis" find lower effects vs. the individual studies that find higher magnitude. In other words- individual studies tend to overstate results. There are many reasons for it, including regression to the mean. |
|
|
|
👉 Do not jump to a conclusion when reading ONE study or book - it is likely to be overstated. The paper also points out (figure 2) that individual studies tend to overstate results vs. combined or meta-analysis. The magnitudes of the individual studies are wider but when we combine many results - the average - naturally shrinks towards the mean. So we should be conservative when reading newspaper articles that report based on a single study.
👉Conduct a deeper investigation and consider behavior modification when many reputed sources agree.
But if many studies over time from reputed sources find the same thing, we should pay a bit more attention. For example, it seems that for beef and bacon (circled in red) a lot of results do point to increasing relative risk of cancer as we can see with all the points being on the right. The disagreement seems more about the magnitude of the association but they are all > 100%.
|
|
|
|
|
Example II. Newspaper headlines on meat vs. cancer
In the following section, we dig deeper into the The Guardian (Oct 26, 2015) headline about processed meat and cancer.
|
|
|
|
Using the first headline from The Guardian (Oct 26,2015) about processed meats we illustrate the process.
👉Step1: Check the quoted authority and its reputation. We circled the authority in red in the sub-headline and it is the World Health Organization (WHO) - a reputed body. So we go to the World Health Organization (WHO) website (and find the Q&A) and check that indeed the article IS coming from them.
Also, WHO (wiki) is NOT funded by a particular lobby but it is funded by various countries - which can apply political pressure - but at least it is a few steps removed from direct corporate funding so we are more likely to believe the source.
👉Step 2: Check if related articles and meta-studies got published in a reputed Journal?
We search for scientific articles related to red and processed meat in the WHO guidelines and the data points cited above to find a meta-study in the Lancet and find that indeed it is a pretty reputed medical journal "among the world's oldest and best-known general medical journals."
👉Step 3: Assess the strength of the evidence
In their excellent detailed Q&A document which is relatively easy to understand, the WHO says red meat is classified under Group 2A, "probably carcinogenic to humans" and processed meat is classified as Group 1 "carcinogenic to humans." The WHO explains that for red meat there is limited evidence hence the probably causing cancer:
"Limited evidence means that a positive association has been observed between exposure to the agent and cancer but that other explanations for the observations (technically termed chance, bias, or confounding) could not be ruled out.
But for processed meat there is stronger evidence that it is carcinogenic to humans (causes cancer):
"In other words, there is convincing evidence that the agent causes cancer. The evaluation is usually based on epidemiological studies showing the development of cancer in exposed humans. In the case of processed meat, this classification is based on sufficient evidence from epidemiological studies that eating processed meat causes colorectal cancer. "
👉Step 4: Fact check the click-baity headline
So what is going on is that the WHO is "equally sure" that both smoking and processed red meat do cause cancer but the amount of cancer the two cause is different! Smoking causes WAY more cancer than eating processed meat. Cancer UK has provided some excellent pictorial illustrations used on the right. Quoting WHO themselves:
No, processed meat has been classified in the same category as causes of cancer such as tobacco smoking and asbestos (IARC Group 1, carcinogenic to humans), but this does NOT mean that they are all equally dangerous.
👉Step 5: Use common sense and other checks to come to a reasonable place.
We go to a few websites like the National Health Service (NHS) UK and see that they recommend cutting down on our red meat portions. We also see that Harvard researchers suggest the same thing. Sabzbag's own meta-study of 50 nutrition books and meta-studies also ranks "lower red meat" as the #4 advice (59% agreement).
But come on, it is not like you really believe eating more cheeseburgers is the way to be healthy regardless of what experts say!
|
|
| |
|
|
|
|
Example III. Investigating Tom Brady vs. tomatoes
The January 6, 2016 Fox news headline focuses on tomatoes and a (potentially) weird belief that we investigate in the section below.
|
|
|
When we actually read the article and follow through the links, the key point it makes is that they are basically following what appears like their own version of the Sabzbag 0 junk - 1 grain - 2 legume (or non - red meat protein) and 3 vegetable - framework:
“So, 80 percent of what they eat is vegetables. [I buy] the freshest vegetables. If it’s not organic, I don’t use it,” says Campbell.
“The other 20 percent is lean meats: grass-fed organic steak, duck every now and then, and chicken. As for fish, I mostly cook wild salmon.”
The chef, who only worked in restaurants before snagging the Brady gig, says the food he prepares for the celebrity family is “very different than a traditional American diet.” When it comes to carbohydrates, only whole grains like brown rice, quinoa and millet are allowed. White flour and white sugar are banned. As is coffee (or caffeine of any kind), dairy and fungus like mushrooms.
Now maybe Tom Brady has an allergy to tomatoes - it could be his own body's quirk or he simply has a weird belief. But that is not the point or the actionable advice, the main effect that is backed by our own meta-studies (90% agreement) is that eating 80% vegetables has got to improve one's nutritional outcomes. And this is backed by common sense - we all know that eating vegetables is good for us!
|
|
| |
|
|
|
|
Our top 7 tips for consuming information
👉 1. Look for reputed meta-studies.
👉 2. Small sample sizes, short study windows, not reputed journals are all red flags.
👉 3. Do not change your diet based on one article.
👉 4. Build up your nutrition knowledge portfolio over time and carefully.
👉 5. Ask yourself, are these people trying to sell me something?
👉 6. Do not just take one doctor's word for nutrition but have an informed conversation.
As our scientific consultant Dr. Richa Jain, an assistant professor of Pediatric Hematology & Oncology, told us - "Doctors get trained in treatment once you have the diseases. We tend not to be nutrition specialists. So, reading reputed meta-studies and asking your doctor questions is likely to help."
👉 7. WhatsApp messages are not Wikipedia! We are all suffering from COVID-19. During such times, we all forward many messages on social media. Since there is no treatment for COVID-19 so far, some companies and individuals prey on the vulnerable using "alternative medicine" as a cover. Please fact check messages instead of blindly forwarding or following them. |
|
| |
|
|
|
|
|
|
Sabzbag Recipe Comics #2: Persian Braised Eggplants with Veggies |
|
|
