#6: Sugar-coated conspiracies: How ‘publication bias’ amplifies half-truths
Hi, this is Samarth. Welcome to The Interval, a fortnightly newsletter to think deeply about the truth-seeking process and forces that distort it. You can read more about it and sign up here. Here is a link to all previous issues. Please send feedback to email@example.com or just reply to this email.
In today's issue: Journalists are hard-wired to look for instances of shadowy wrongdoing. This is especially true for stories that involve cronyism, conflict of interest or payoffs in scientific research and/or regulatory policy. But a New York Times story that called out a prominent Indian scientist—overseeing packaged labelling in India—shows how this bias can lead to bad reporting and misinformation. Even when there is smoke, there isn’t always fire. It's a 15-minute read.
Imagine a hypothetical scenario. Journalists in five different newsrooms start chasing a story. An activist alerts all of them to a potential case of corruption: a senior bureaucrat might be tweaking policy to favour a real-estate developer. The reporters start digging.
One finds connections: the duo go for a golf outing every month, their kids study in the same school, and travel records show both holidayed in Spain last summer. The newspaper splashes this scandalous ‘scoop’ on its front page.
The other four reporters have the same information but their editors want more. “This only suggests these guys are friends,” says an editor “but it does not show if and how the rules are being tweaked to favour this guy and what happened inside the government. Show me that. Get me the evidence.”
They continue digging. They interview several people. They access internal government records. But they can’t find any solid evidence. Instead, the opposite happens: they find holes in the corruption hypothesis. The bureaucrat had rejected proposals that favoured his friend. He was following due process.
“This means there is no story here. It’s just hearsay,” says the editor. “Keep an eye out for potential leads. But drop this for now. Go work on something else.”
Journalists move on. End of story.
Look what just happened here. Five reporters, from five newsrooms, worked on the same story, but the public gets to know about one. They only know of one because the other four did not publish. They only got to read the accusatory story establishing friendship. They may even laud the ‘courageous reporter’ and the ‘gutsy publication’ to expose the ‘state-capital nexus’.
But the contradictory findings of the other four remain an internal record because “we did not find the bureaucrat favoured his friend” is not considered news. That is normal — and normal things don’t make news.
But the fact is that journalists worked on it and found ‘nothing’ — and ‘nothing’ is a finding. It is what I call a ‘null story’.
Null stories are not appealing: the reporter and the publication have no incentive to say ‘we know there is a problem but we don’t know who is behind it and all we know is the story of that other guy is misleading.’
This ‘publication bias’ of not carrying null stories is a journalistic blind spot: it allows half-baked claims to circulate in the public discourse which distorts our understanding of complex issues, gives birth to conspiracy theories, and diverts our attention away from what’s really happening in the world.
This is especially so in stories involving money, power and politics.
In today’s edition of The Interval, I will tell you a ‘null story’ on India’s food politics featuring a shady research organisation, a scientist, dubious statistics and a propaganda campaign to promote sugar. And I will make a case for publishing null stories. It’s not just about media critique: refutations are fundamental to the truth-seeking process.
I. India’s labelling policy and NYT report
In September 2019, the New York Times published a bombshell report: “A Shadowy Industry Group Shapes Food Policy Around the World!”
It begins with a discussion on India’s progressive front-of-pack labelling regulation. (It was introduced to solve the problem of misleading “healthy food” labels I had discussed in the previous issue.)
The idea of the regulation is simple. From a story I wrote for Mint:
If the content of fat, salt or sugar—the three nutrients of concern—exceed a specified quantity, the food packet would come with a front-of-pack warning label. An easily understandable red dot for each element that surpasses a defined threshold.
The front-of-pack label, its advocates believe, would empower consumers in making informed choices and incentivise companies to innovate in food technology to bring down fat, salt and sugar levels.
The regulation is stuck for around six years now following fierce opposition from the food industry.
The NYT report offered a clue for these delays: the appointment of a specific scientist to shape this regulation.
Here is what happened: The first draft of this policy came out in 2018. Industry executives expectedly expressed concerns. So, the FSSAI, India's food regulator, set up a three-member expert panel to consult stakeholders and propose revisions.
This panel was headed by Dr Boindala Sesikeran. This, the NYT told us, was a problem. Here are the first three paragraphs:
When the Indian government bowed to powerful food companies last year and postponed its decision to put red warning labels on unhealthy packaged food, officials also sought to placate critics of the delay by creating an expert panel to review the proposed labeling system, which would have gone far beyond what other countries have done in the battle to combat soaring obesity rates.
But the man chosen to head the three-person committee, Dr. Boindala Sesikeran, a veteran nutritionist and former adviser to Nestle, only further enraged health advocates.
That’s because Dr. Sesikeran is a trustee of the International Life Sciences Institute, an American nonprofit with an innocuous sounding name that has been quietly infiltrating government health and nutrition bodies around the world.
Sesikaran’s presence in the expert committee, health advocates told the NYT, is not acceptable: “To have a covert food lobby group deciding public health policy is wrong and a blatant conflict of interest,” one activist told the publication.
The report then dives deep into the history of the International Life Sciences Institute (ILSI): it “championed tobacco interests during the 1980s and 1990s”; it’s “China affiliate helped shape anti-obesity education campaigns that stressed physical activity over dietary changes, a strategy long espoused by Coca-Cola”; studies showing how members strategised the need to “step up their fight against the W.H.O.’s increasingly tough stance on sugar.”
This shady track record of ILSI suggested that Dr Sesikaran is up to something.
Down to Earth, a fortnightly magazine published by the New Delhi-based Center for Science and Environment — whose director was quoted in the NYT story and whose members were consulted on this policy — published an article which said the “report of the 2018 committee, headed by Sesikeran, has never been made public. But it is this committee which brought in the most industry-friendly changes.”
I was intrigued. What is in that report? How exactly is Sesikeran tweaking the rules and whom will it benefit?
None of this information was available. So I filed a request under India’s Right to Information Act to get records of these expert group meetings to find out what Sesikaran had done.
II. “Things are not what they seem on the surface”
What did the documents reveal?
Sesikaran’s group met thrice between September and December 2018. In the first meeting on September 7, the group introduced a few changes. From my Mint article:
The changes they introduced included modifying the ingredients that would determine what is an HFSS food: in the new definition, high sugar would refer to ‘added sugar’, not ‘total sugar’; high fat would focus on ‘saturated fat’, not ‘total fat’; and salt was replaced with sodium. Limits were modified accordingly.”
Reasons were given for this change but critics said these were done to please the industry. It’s debatable.
But two things in those documents caught my attention.
One: While activists pointed out that changing ‘total sugar’ to ‘added sugar’ on the label was an industry-friendly move, meeting minutes revealed otherwise.
In the October 8 2018 meeting, industry representatives opposed what Sesikaran’s group had proposed. They wanted ‘total sugar’ as a mandatory declaration and ‘added sugar’ as voluntary.
Two: In the meeting on 27 December 2018, the representatives of the Indian Sugar Mills Association came in as special invitees to make a case against the labelling policy.
The association has not come across any report regarding the adverse effects of sugar consumption, they told the panel. Sugar is an essential part of the diet, an inexpensive source of energy, and India’s per capita consumption of sugar is the lowest in the world. So putting a red mark on the pack — which will discourage the consumer from consuming it — for sugar levels beyond a threshold is not a good idea, they said.
Meeting minutes show that Sesikaran countered them. Most people do not have the capability to read the labels, he told the sugar lobby, and colour coding matters because it is easy for consumers to understand. The thresholds were based on WHO standards and they were not targeting sugar in isolation — fat and salt were part of the regulation, too.
Another scientist in the panel made it clear: “systematic data is available regarding excessive sugar consumption and its adverse health effects”.
Information from these government records complicates the NYT and Down to Earth narrative. While I can’t — and I don’t want to — prove that Sesikaran is an innocent scientist, it does suggest he wasn’t always behaving like some food industry automaton. He actually showed some independence.
More contradictions appear in the Sesikaran narrative.
III. The failed PR campaign: “You are not eating enough sugar”
In October 2020, nearly two years after the Sesikaran meeting, India’s sugar lobby (ISMA) started a bizarre public relations campaign. The association launched a new portal — meetha.org — to fight the ‘misinformation surrounding sugar’.
India's union food secretary was present at the website launch.
On a Zoom call, the association’s director-general Abhinash Verma told me the origin story.
I have been attending a lot of conferences and meetings at the international level and we are concerned about the campaign that is going on against sugar to make everybody believe that if you reduce your sugar consumption you will be the healthiest person in the world. This campaign is getting bigger and bigger in India.
Someone is behind this campaign without much research. We realized that none of these scientists, anywhere in the world, have ever concluded that sugar is bad. Nobody, be it doctors or nutritionists will tell you that consuming sugar per se will lead to diabetes or dental caries or will increase your chances of getting cancer.
So as a responsible association of sugar producers, we have started a small campaign to discuss in a public forum what is correct and what is incorrect. Our meetha.org could be the platform to convey the right details and fill the vacuum.
Verma spoke to me for close to an hour. He clarified: they are not asking people to eat more sugar. They are asking for moderation and busting misconceptions which are leading to misguided policies like front-of-pack labelling.
I wrote about sugar in The Interval's previous issue and I won't get into the specifics here. As Daniel Engber pointed out in a fine Atlantic essay:
Just as the history defies a simple reading, the research on nutrition—ample and diverse though it’s been—isn’t close to dispositive. We can’t prove the case against sugar, and we can’t prove the case against that case, either.
This campaign has terribly failed which is why you have probably never heard about it. Their YouTube channel has only four subscribers and their Twitter account has less than 200 followers.
What matters for this discussion is the sugar association’s stunning claim: Indians are not eating enough sugar.
The per-capita consumption of sugar in India is 19gm per day, they said, which is much lower than the recommended 30gm/day by ICMR and 50gm per day standard by WHO.
If sugar consumption is so controlled, what is the fuss about?
These numbers come from a government survey done by the National Institute of Nutrition. But they are hard to believe and they don’t stack up with other data sources I checked.
Critics soon found a hint: ILSI was the partner in this survey, suggesting that the numbers could possibly be fudged to promote the case made by the sugar industry.
This is possible. Let’s say ILSI influenced the study to get the numbers favouring the sugar industry.
But now, go back to those closed-door government meetings where lobbyists have the most power to skew things. And there, Sesikaran — ILSI’s own trustee — had rejected these bogus arguments of the sugar lobby. The fact that their case was not accepted by the food regulator panel was partly the reason they launched this PR campaign.
So which theory should we accept?
The claims against Sesikaran just don't add up.
IV. "The wrong target"
The questions that emerged from my reporting got me thinking: Sesikaran, perhaps, may not be the villain the press is telling us he is.
His view on this whole controversy was not publicly available. So I called him this week to understand what he had to say.
“If what I said in the committee meetings is made public, everyone will know fully if there was any kind of hanky-panky in the committee's decision. It should have been transparent for the public to know,” he told me.
Soon after the New York Times story, Sesikaran wrote to the FSSAI: “Don't be hesitant. If you say, I will step down. I have been on several committees. You know my track record. If there is anything that indicates that I have a bias, you can always haul me up or if you want me to leave the FSSAI committee, I can do that too,” he said.
The FSSAI didn't respond or say anything. So he continued to attend a few more meetings after that (he was a member of multiple committees). But that was the time his term in the FSSAI was about to end (scientific member participants have to periodically renew their terms). At the time of renewal, he did not send his name, nor did the FSSAI ask him to.
“That's how it ended,” he said.
“The industry has direct access to FSSAI,” he told me. “They have their own channels, and they have been working through those channels. ILSI has never represented on behalf of the industry.”
On the phone, Sesikaran made a strong case for better labels. “The industry needs to put the front of pack labels and they need to provide more information. There is no going back or second thought on this,” he said.
“It is quite disappointing. We spend our entire lives setting dietary guidelines and RDAs and all that and just on the basis of an association with another thing people are making these comments about me,” Sesikaran said.
But let's be clear here: Another expert who had worked on the labelling policy told me there is no doubt that the food industry has a deep-rooted influence inside India’s regulator. But Sesikaran is the wrong target. “It’s been close to three years since he is out of this panel, and nothing has moved. How do you explain that?”
“As food scientists, we work with the industry and we work with the government. And we offer our expertise. Don’t cast aspersions on us without evidence.”
V. Why this matters
I took what I found to David Merritt Johns, a public health historian at the Yale School of Public Health.
His excellent 2018 paper in Science — debunking a manufactured “sugar conspiracy” — with Gerald M. Oppenheimer partly pushed me to write this ‘null story’.
From their article in Slate:
Over the last two decades a remarkable shift in focus has taken place in the nutrition field—we’ve gone from blaming fat for our expanding waistlines to fingering sugar instead.
And who was to blame for our nutritional befuddlement?
In late 2016, there emerged an explanation that appeared to cut through the fog: It was the sugar industry. Researchers with the University of California announced they had unearthed secret archival documents showing that in the mid-1960s, the industry-backed Sugar Research Foundation had covertly paid top scientists at Harvard to conduct a literature review playing down the role of sugar in heart disease and pinning the blame on dietary fat instead.
But as we detailed last month in Science, our own examination of the historical events in question shows this alluring tale of industry meddling is based on a highly selective and profoundly flawed interpretation of the history.
The long-deceased Mad Men–era Harvard scientists who stand accused of having been “paid off” to “shift the blame” to fat were, in fact, already on record in support of low-fat diets as a way to fight heart disease for nearly a decade before the sugar men came calling. In adopting this stance they were in sync with the dominant nutritional paradigm of the era: the idea that the fatty American diet, by raising cholesterol levels in the blood, was behind the epidemic of heart attacks that was killing so many middle-aged breadwinners. No blame-shifting was even required!
So what explains the emergence of this shadowy narrative about Big Sugar? Certainly there is a rich tradition in contested areas of science of using bits and pieces of history to add urgency to one’s claims.
The Sesikaran-ILSI saga had similarities with this case. So I emailed David to get his views on what I had found in my reporting.
Here is our email exchange, slightly edited for clarity:
Samarth: What do you think about the journalistic adage of “follow the money”? This is what editors tell their reporters looking to expose corruption. While this connection can be informative, it can sometimes be stretched too far. Just a hint of a funding connection from X to Y is considered enough to explain the behaviour of Y. In my own reporting, I have seen that it is a good starting point but it is often not the full story. It appears that your research on the nutrition wars is suggesting the same. What do you think?
David Johns: "It is always a good idea to follow the money. Money makes things happen. And very often things don’t or cannot happen in the absence of money.
That said, there is no question that it has become a popular academic and journalistic trope to seek out financial connections and then use those ties to discredit some public figure, or to use those ties to explain why one (bad) thing happened while another (good) thing didn’t."
Samarth: What are its implications for our understanding of complex issues in food policy? What do we miss out on?
David Johns: "One relevant point in the journalistic realm is that I think there has been a tendency to focus on and demonize particular individual scientists who have had or do have relationships with industry, rather than consider how the industry’s activities as a whole might tend to amplify certain ideas and arguments that align with the industry’s interests.
If the conferences are set up in such a way that they amplify certain industry-friendly messages (e.g., they promote exercise instead of denigrating junk food), then maybe you have academic articles are published as a result of the conference. Maybe journalists cover the conference, and they transmit that message.
So the industry can influence the conversation and shape the science and policy even if the scientists themselves are not at all shills of the kind you would imagine from the movies (i.e., slimeballs who take the industry money and then do the industry’s bidding without regard or whether it’s right or wrong).
Industry achieves its goals simply by amplifying the part of the story that aligns with its interests.
It’s true that the scientists who play along may be abetting the industry in ways they don’t even realize; but that’s different from suggesting that X thing happened because of the industry’s influence (e.g., “the sugar industry paid off Harvard scientists and that’s why we got the low-fat diet!”)"
Samarth: So in the simple “follow the money” mental model, the scientist is guilty. But just because I can’t show that he has done something wrong, that doesn’t mean he has not done anything wrong. Maybe I am a bad reporter who was not able to uncover what he had done in the background?
So one can say: sure, their evidence is weak, but pointing out that someone else’s evidence is weak may be read as a defence of that person — which is not what I want to do. I wonder if you thought about this while writing your paper?
David Johns: "Yes, we were very careful for the reasons that you mention not to suggest that we knew for sure that the sugar industry wasn’t more influential and nefarious behind the scenes than we realized or had been able to discover in our archival research. But the point is that you need to have evidence for such things before you can make strong claims."
David and his co-author aptly summed up the problem in their Slate piece:
But conspiratorial tales, when not grounded in strong evidence, can pose a real danger to our ability to make good public health policies and understand how science actually works. Stories about clandestine payoffs and corporate cover-ups by definition suggest that some scandalous truth was being hidden...Avoiding these traps requires a commitment to evidence and an appreciation for the complexity of scientific research and its history.
That is my point — a point I believe is worth endlessly repeating. We must hold the powerful accountable and we must investigate the influence of money in politics. But we should do that with a commitment to evidence and the truth.
Informed refutations are a crucial part of the truth-seeking process and journalists can contribute to it by finding a way to publish their ‘null stories’.
The Interval Newsletter
Join the newsletter to receive the latest updates in your inbox.