First Draft case study: Exploring the controversy around Dengvaxia and vaccine misinformation in the Philippines

This article is part of a series on health misinformation.

WARNING: This study continues images and themes some readers may find distressing. Find out more about dealing with vicarious trauma.

Dengue is a viral infection found in tropical and sub-tropical climates, and is currently the most rapidly spreading mosquito-borne viral disease in the world. During the past five decades, the global incidence of dengue has risen 30-fold, with the disease now endemic in more than 100 countries.

In April 2016, Dengvaxia was introduced in the Philippines via a school-based campaign organized by the Department of Health (DoH). The decision was influenced in part by World Health Organization (WHO) guidelines. At the time, the WHO guidelines recommended the use of Dengvaxia in areas where “dengue is highly endemic as defined by seroprevalence in the population targeted for vaccination.” The same vaccine had also been licensed previously in 20 other countries around the world, but not distributed. During this vaccination campaign, roughly 830,000 children aged 9 and over received at least one dose of the vaccine.

Prior to its introduction, the use of Dengvaxia was already a topic of debate within a small portion of the Philippine medical community. Various figures opposed the vaccine, citing a contentious theory championed by scientist Scott Halmstead called antibody-dependent enhancement (ADE). According to the theory, if a child has never had dengue before, Dengvaxia could possibly make a dengue infection more severe than it would have normally been without the vaccine.

This small group of medical professionals, led by University of the Philippines clinical epidemiologist Antonio Dans, took to Facebook in 2016 to highlight their concern and campaign against the government’s introduction of the vaccine.

Soon after its distribution in the Philippines, Dengvaxia became an even bigger subject of controversy. In November 2017, Sanofi, the French multinational pharmaceutical company that developed the drug, released an updated analysis on Dengvaxia suggesting that administering the vaccine to seronegative children—or those children who have never been exposed to dengue—could increase their risk of developing a more severe form of dengue.

Dengue vaccinations were officially suspended in the Philippines shortly thereafter. The decision to shut down the vaccination program seemed to confirm what medical professionals skeptical about Dengvaxia had already been voicing online. Furthermore, this policy U-turn set off a moral panic among parents and the public at large who believed the government, and specifically the DoH, had put their children at risk by administering a dangerous drug.

In part, this incident led to a rapid decline of vaccine confidence among Filipinos. The consequent politicization of Dengvaxia, including an online and offline crusade against the DoH led by individuals from the Public Attorney’s Office (PAO), in tandem with the mainstream media’s amplification of rumors and publication of highly emotional content such as videos depicting children “dying” from Dengvaxia, did little to assuage public panic.

A lack of any clear consensus from the medical community in the Philippines only encouraged the cacophony of voices and claims in online spaces that Dengvaxia was dangerous, allowing conspiracy theories buoyed by #denggate—a popular hashtag used to associate misinformation about Dengvaxia with dengue—to proliferate online.

In this case study we outline the main themes of the social posts that were shared most prominently, highlight some of the key accounts spreading misinformation, and discuss the impact of the intersection between the Dengvaxia controversy and vaccine hesitancy more generally.

TIMELINE OF KEY EVENTS

Fig 3: Dengvaxia Timeline Seronegative subjects refers to people who had not been exposed to the dengue virus infection prior to vaccination with Dengvaxia.

Loss of Vaccine Confidence

According to the WHO, “vaccine hesitancy” is a delay in acceptance or refusal of safe vaccines despite the availability of vaccination services. The organization has classified vaccine hesitancy as one of 10 big threats to global health in 2019.

Today, one in five children worldwide fail to receive routine immunizations, and about 1.5 million children die each year of diseases that could be prevented by vaccination. Concern over vaccine safety is one of the most dominant reasons for vaccine hesitancy.

Research by the London School of Hygiene & Tropical Medicine (LSHTM) suggests that the highly politicized response to the campaign against Dengvaxia was the primary reason for increased vaccine hesitancy and public mistrust within the Philippines. A large part of that politicization resulted from a protracted Senate investigation into the Dengvaxia debacle. The long Senate hearings were televised widely and focused more on corruption allegations than the safety of the vaccine, making space for conspiracy theories around financial kickbacks and malfeasance to grow. And by leaving the safety of the vaccine unaddressed, the hearings created a vacuum that was filled by conspiracy theories and misinformation about how the vaccine was linked to numerous children’s deaths.

Ultimately, the investigation gave Persida Acosta, the Philippines’ chief public attorney, a platform from which to launch a highly publicized and uncorroborated crusade linking Dengvaxia to the deaths of hundreds of Filipino children — further sullying the reputation of the DoH.

According to a study by the Vaccine Confidence Project at LSHTM, in 2018 fewer than a third of Filipinos strongly agreed that vaccines were important, down from 93 percent in 2015. Since the Dengvaxia controversy, the confidence in vaccines among Filipino parents plummeted from 82 percent in 2015 to only 21 percent in 2018.

Fig 1: London School of Hygiene & Tropical Medicine Infographic showing drop in vaccine confidence in Philippines Credit: LSHTM

While it is difficult to quantify, a growing number of medical professionals believe that the Dengvaxia crisis eroded trust in vaccines overall and contributed, to a certain extent, to the recent spike in polio and measles in the Philippines.

Interest in the Philippines in Dengvaxia over Time

In addition to the study by the LSHTM, an analysis of Google Trends data shows a spike in Dengvaxia-related search queries by people located in the Philippines between November 2017 and March 2018, which is when the majority of the online misinformation about the vaccine was being shared.

Fig 2: Source First Draft Google Trends Data

 

BACKGROUND

While it is easy to be critical of the online conspiracies that circulated widely during the Dengvaxia scandal, it is important to recognize that the manufacturers of the vaccine (Sanofi) and the WHO have acknowledged that there are risks associated with Dengvaxia. As those of us that study misinformation know, the most effective type of misinformation is that which includes a kernel of truth. In this case, the official warnings about Dengvaxia fueled further speculation and rumors online.

Sanofi’s official review, conducted in late 2017, found that in rare cases Dengvaxia can have a potentially dangerous, unintended consequence: If people who never had dengue are vaccinated and later become infected, the vaccine may provoke a much more severe form of the illness.

When the WHO Global Advisory Committee on Vaccine Safety considered Sanofi’s results, it concluded that there was an increased risk of severe dengue disease in people who had never been infected.

As dengue is endemic in the Philippines, the risk of this side effect impacts about 15 percent of vaccinated individuals. Of this seronegative 15 percent (those who have never had dengue), only around four in 1,000 patients vaccinated were found to have developed severe dengue disease within five years of observation.

Based on its analysis, the WHO issued a conditional recommendation, advising that Dengvaxia be used only on populations that had been previously infected with dengue virus.

The DoH suspended the dengue vaccination program and the Philippine Food and Drug Administration (FDA) announced that it would stop the sale of Dengvaxia in December 2017 until Sanofi updated its packaging label to reflect new warnings about the drug. The events ultimately triggered two congressional inquiries and a criminal investigation into the vaccine in the Philippines. In February 2019, the Philippines revoked Sanofi’s product licence for Dengvaxia, permanently halting the sale, distribution, and marketing of the dengue vaccine in the country. The FDA said that as of December 2018 Sanofi had failed to comply with post-marketing authorization requirements in place to assure that a drug remains safe to sell.

Concepcion Yusop, a national immunization program manager, shows an anti-dengue vaccine Dengvaxia inside a vaccine storage room in Sta. Cruz city, Metro Manila, Philippines December 4, 2017. REUTERS/Romeo Ranoco

Concepcion Yusop, a national immunization program manager, shows an anti-dengue vaccine Dengvaxia inside a vaccine storage room in Sta. Cruz city, Metro Manila, Philippines December 4, 2017. REUTERS/Romeo Ranoco

The Philippine FDA determines whether products such as drugs, vaccines, food, and cosmetics are safe for the public. The FDA issues a certificate of product registration (CPR) when a product can be sold in the country. Once a company’s product is in circulation, the company is required to submit several auxiliary documents—part of the post-marketing authorization requirements—demonstrating that the product is still safe. On December 22, 2015, the FDA issued its CPR to Sanofi. Yet, Sanofi failed to submit the necessary post-marketing requirements after being allowed to sell Dengvaxia.

Sanofi disagreed with the FDA’s findings. In March 2019, the Philippine Department of Justice said it had found probable cause to indict officials from Sanofi, as well as former and current Philippine health officials, over 10 deaths it said were linked to use of the dengue vaccine.

Past skepticism among health professionals and researchers

It’s crucial to note that the initial rollout of Dengvaxia was met with some skepticism from researchers, scientists, and medical health professionals. According to Scientific American, Antonio Dans and pediatrician Leonila Dans, both clinical epidemiologists at the University of the Philippines Manila College of Medicine, had concerns about Dengvaxia.

In March 2016, along with other medical professionals, they wrote to then-Secretary of Health Janette Garin warning that the vaccine could be risky for some children and that the Philippines might not possess enough trained health care workers to monitor them for possible adverse effects. Scientist Scott B. Halstead also very publicly let the WHO know about his concerns. In a December 2016 paper in The Journal of Infectious Diseases, he disputed a claim by the WHO’s principal advisory group on vaccines that the risk of hospitalization for children aged two to five peaks in the third year after vaccination and then “dissipates”. For those children with no prior dengue infection, Halstead suggests instead a new candidate vaccine that the US National Institute of Health is testing on 17,000 people in Brazil.

CONFLICTING PERSPECTIVES

The exact number of individuals who have died as a result of Dengvaxia is uncertain and a point of contention among different groups. According to the WHO, an increasing number of Dengvaxia-related AEFI (Adverse events following immunization), including deaths, were reported, after which the Dengvaxia Investigative Task Force (DITF) was established to review deaths following dengue vaccination. Pediatrician and panel member Juliet Sio-Aguilar, from the University of the Philippines-Philippine General Hospital (UP-PGH), said the team was recommending further studies because it was difficult to directly connect the three deaths reported in February 2018 to Dengvaxia. In an interview we conducted in July 2019 with Heidi Larson, director of the Vaccine Confidence Project at LSHTM, she said there is no evidence of a causal link between Dengvaxia and deaths in the Philippines.

This is in stark contrast to the message being promoted by particular entities within the Philippine government, including the Public Attorney’s Office (PAO). Led by Persida Acosta, the PAO reported in February 2017 that more than 8,400 people had become sick and more than 130 had died from Dengvaxia.

The Department of Health said in a statement in April 2018 that from March 2016 to March 2018 a total of 3,281 students were hospitalized in different public and private hospitals for “various illnesses” and that “65 deaths were also reported which are now under investigation.” The statement later noted that over 98 percent of those hospitalized were eventually sent home as false alarms. The DoH never confirmed that any of the children’s deaths were caused by Dengvaxia.

People display signs and a mock syringe, with the phrase "3.5 billion pesos Dengvaxia fund investigate" featured on it, during a protest in front of the Philippine Department of Health (DOH) in metro Manila, Philippines December 5, 2017. The poster (L) reads: "Sanofi and DOH should be held accountable". REUTERS/Romeo Ranoco

People display signs and a mock syringe, with the phrase “3.5 billion pesos Dengvaxia fund investigate” featured on it, during a protest in front of the Philippine Department of Health (DOH) in metro Manila, Philippines December 5, 2017. The poster (L) reads: “Sanofi and DOH should be held accountable”. REUTERS/Romeo Ranoco

The PAO, supported by various groups on Facebook, now claims that hundreds of children have died as a result of Dengvaxia. Much of the evidence provided by the PAO linking Dengvaxia to the deaths of children came from a team of PAO forensic investigators led by Erwin Erfe. The team performed autopsies, sometimes in front of live television crews, which supposedly linked the fatalities to Dengvaxia. However, this team of PAO investigators was heavily criticized and dismissed by a large part of the Philippine medical community, including the Philippine Society of Pathologists, as well as Antonio and Leonila Dans, among others, for being unqualified as forensic specialists.

These conflicting narratives and perspectives have provided ammunition for the conspiracy theorists online, allowing them to peddle emotive and sensational misinformation which proliferated across social media.

AMPLIFICATION: SOCIAL MEDIA

An initial phase of qualitative research, including a review of newspaper pieces, academic journals, and social media, revealed a series of keywords and hashtags related to Dengvaxia and the ongoing vaccine crisis in the Philippines. These included #dengue, #fightthebite, and #vaccines. However, because of their nonspecific nature — the hashtags have been used in multiple countries — they were deemed too general for tracking social media activity related specifically to dengue in the Philippines.

The major Dengvaxia panic that erupted in the Philippines at the end of 2017, which captured headlines in the country and around the world, meant that the term “Dengvaxia” (and its associated hashtag #dengvaxia) was strongly linked to Filipino users of social media. We thus chose Dengvaxia as the most effective keyword to start our searches.

To analyze Facebook data, we used the Historical Data feature of CrowdTangle (a social media monitoring platform), inputting the keyword “Dengvaxia” as the search term. We pulled all Facebook posts which included that keyword from December 2015 to July 2019. We went back as far as December 2015 because that was when then-President Aquino made the decision that the Philippines would be the first Asian country to approve the commercial sale of Dengvaxia.

The resulting dataset contained all posts which included “Dengvaxia” from every Facebook Page in CrowdTangle’s archive. In total, the dataset held 78,157 different posts from 4,965 Facebook Pages. The posts produced 19,854,183 interactions during this time period.

Using Python and Pandas (a data analysis package for Python), we examined the top posts during the two peak interaction times between November 2017 and April 2018, and between January 2019 and April 2019. This analysis also produced the top Pages/accounts posting information about Dengvaxia, as well as those Facebook Pages that produced the most total interactions. Using this data, we were able to identify which accounts were dominating the online narrative around Dengvaxia and what kind of content they were producing.

To analyze Twitter data, we searched the term “Dengvaxia” on the platform during the same time periods we had for Facebook. The analysis allowed us to see tweets that included Dengvaxia in the text, as well as the total retweets, likes, shares, timestamps, and hashtags associated with those tweets. Only tweets were included in the analysis (retweets were not included in the dataset).

AMPLIFICATION: HEADLINES

The Google Trends team in London provided us with a dataset of headlines and news stories published in English, and related to dengue and Dengvaxia from January 1, 2016, to September 1, 2019. Using Python and the Python package Natural Language Toolkit (NLTK) — a suite of libraries for natural language processing — we analyzed the keywords and phrases in these headlines.

Common English words such as “to,” “for,” and “is” were removed from the language corpus during pre-processing to refine the language for analysis. Stemming (removing suffixes), lemmatization (working with the roots of words), and basic machine learning techniques such as vectorization were used to return 1) a set of individual key words, 2) bi grams (combinations of two words), and 3) tri grams (combinations of three words) that appeared most frequently in the headline corpus.

Because the analysis looked only at headlines published in English, it misses the many news stories and headlines published in the various languages used in the Philippines. Therefore the results shouldn’t be taken as a comprehensive examination of news narratives around Dengvaxia in the Philippines, but as an auxiliary analysis which supports other findings in the study.

TYPES OF MISINFORMATION ONLINE

WARNING: This section contains two images of a human brain.

Unsubstantiated death tolls

There are many posts and memes online claiming Dengvaxia caused the deaths of hundreds of children, which were supported by Persida Acosta’s campaign and Erwin Erfe’s controversial autopsies. Yet to date, there has been no causal link connecting the deaths of children in the Philippines to the Dengvaxia vaccine.

Fig 4: An Instagram post originally authored by the user @truthinjection whose account had approximately 1,500 interactions before it was removed. While active, the page contained numerous posts with anti-vaccine content.

Fig 5: An Instagram post originally authored by the user @bye_bye_big_pharma whose account has approximately 13,500 followers and contains numerous posts with anti-vaccine content.

The memes amplify the controversial findings from Erfe’s autopsies, which attempted to connect the deaths of hundreds of children to Dengvaxia—the results of which were challenged by broad segments of the Philippine medical community. Many of the memes also carry the hashtag #denggate.

Fig 6: A post on Facebook mentioning both #Denggate and Dengvaxia. The post was shared by Thinking Pinoy, a community page with over 1.3 million likes on Facebook. The post was uploaded on January 9th 2018 and received 14,000 reactions, 9,800 shares, and 1,600 comments.

The above meme, which was produced by Thinking Pinoy and uses the hashtag #denggate, produced over 25,000 interactions, including nearly 10,000 shares. It’s one example of the many other memes that amplified the same misinformation.

Fig 7: A post on Facebook arguing that Dengvaxia led to the deaths of children. The post was shared by Rodrigo Durterte News, a community page with over 315,000 likes on Facebook.

Various other Facebook posts also used Erfe’s findings to augment the narrative that Dengvaxia led to the deaths of children, often linking to articles that no longer exist online or whose sites have been taken down.

Online conspiracies and rumors

In this case study we wanted to explore online misinformation to understand whether it may have played a role in the increase in vaccine hesitancy in the Philippines. Our qualitative research demonstrated that a handful of popular bloggers in the country have been producing and amplifying a steady flow of conspiratorial content and rumors related to Dengvaxia on both Facebook and Twitter — see examples here and here. The posts have, among other things, accused the DoH of bribing families to drop charges related to the Dengvaxia crisis. This content villainizes the DoH and has likely further eroded public trust in the institution. Persida Acosta’s ongoing persecution of the DoH, with the backing of the Public Attorneys Office and some ex-DoH employees, has potentially supported these bloggers’ claims in the eyes of the public.

Blogger RJ Nieto, who posts under the name Thinking Pinoy, was the biggest driver of interactions in the Crowdtangle dataset used for this analysis, with his Dengvaxia Facebook generating 2,887,578 total interactions. Thinking Pinoy and another user, Sass Rogando Sasot, were key players in amplifying the debate around Dengvaxia on Facebook and Twitter. They were also chief drivers of the hashtag #denggate, producing 2 million and 566,000 interactions on Facebook respectively, between December 2017 and April 2019.

On December 3, 2017, Thinking Pinoy published the following meme on his Facebook Page accusing Aquino-era Health Secretary Janette Garin and Sanofi of attempted genocide in relation to the Dengvaxia vaccine, asking: “Is genocide ok if done discreetly?” The Facebook post received over 45,000 interactions and over 15K shares.

Fig 8: A meme was posted on Facebook accusing Aquino-era Health Secretary Janette Garin and Sanofi of attempted genocide. The post was shared by Thinking Pinoy, a community page with over 1.3 million likes on Facebook. The post was uploaded on December 3 2017 and received 32,000 reactions, 15,000 shares, and 3,800 comments.

The post included a link to an article on Thinking Pinoy’s blog, which describes how Janette Garin “chronically ignored expert opinion and launched what possibly is the biggest Public Health scandal in the country.” The blog post on thinkiningpinoy.net gleaned over 100,000 interactions across social media, according to the CrowdTangle plugin. It was also shared by MOCHA USON BLOG, another popular Filipino blogger with over five million followers, as well as other influential Facebook bloggers such as For the Motherland – Sass Rogando Sasot.

Thinking Pinoy also continued to push out content criticizing politicians linked to the Dengvaxia scandal. In one post he wrote, “If he [Bam Aquino] is reelected, he will have 6 more years to protect the interest of Sanofi and his party that benefited from Dengvaxia.”

The politicization of Dengvaxia, and its use as political leverage by people such as Senatorial aspirants Doc Willie Ong and Larry Gadon, is more present during the second spike of social media interest in Dengvaxia we observed between January and March 2019.

Conspiratorial content around how the DoH handled the Dengvaxia vaccine continued to proliferate even after the Philippines issued a national alert around dengue fever in July 2019. Stories such as Dengvaxia’s missing millions exemplified this kind of content:

Fig 9: An image from a Daily Tribune article entitled, “Dengvaxia’s Missing Millions.” The article states that millions of government funds have gone missing, were wasted or are unaccounted for due to the rollout of Dengvaxia. The article also states that the drug was “a scam that coincided with President Benigno Aquino III’s political ploys to be reelected.”

As a result of Persida Acosta’s crusade against the DoH, misinformation was legitimized and further amplified. Acosta’s campaign had support from various Facebook Pages, including Friends of the Public Attorney’s Office – PAO and United Dengvaxia Victims. The pages provided an onslaught of misleading content, including, among other things, the highly contested forensic results collected by the PAO.

Health hoaxes, fake cures, and scams

Another key theme to emerge online was supposed remedies and preventative measures against dengue fever that have been debunked by all medical health authorities.

A Facebook post showing a supposed hospital notice from Sri Lanka that detailed how coconut oil can prevent dengue fever appeared on Philippines Report, a popular Facebook Page with nearly 100,000 followers. The post received over 5,000 interactions. The post is a hoax that has appeared in other countries, such as India, where it was debunked three years ago by Hoax or Fact and this year by Boom.

Fig 10: A flyer alleged to be from a doctor in Sri Lanka was posted on Facebook urging people to use coconut oil below their knees as an “antibiotic” to prevent dengue fever. The post was shared by Philippines Report, a community page with over 163,000 likes on Facebook. The post was uploaded on July 3 2019 and received 4,900 shares.

A Facebook post advocating the use of tawa tawa tea for an assortment of ailments including dengue generated 725 total interactions. Many of those who commented on the post were supportive of this treatment, thanking the author for posting the content. Another example from ArtikuloHealth News also champions the use of tawa tawa to treat dengue. The post produced 1,685 total interactions. A post on July 26, 2019, from the Facebook Page Kalusugan ay Kayamanan (which translates as “Health is Wealth”), also endorsed the use of herbal remedies including papaya and tawa tawa to treat dengue. The post produced 1,115 total interactions.

Fig 11: Posts from various health-related Facebook pages detailing different types of herbal remedies for dengue fever, none of which are a cure for the disease.

 

KEY FINDINGS

There was significant overlap between Facebook and Twitter data. For example, the same spikes in posts occurred around Dengvaxia during similar time periods for both platforms, just as some of the same individuals were controlling the overall narrative and driving much of the debate on Facebook and Twitter.

Conducting an analysis of both platforms also revealed a few differences and nuances. Francis Cruz and Persida Acosta, both key political figures responsible for spearheading much of the backlash against the Department of Health, are relatively invisible on Facebook. Yet, on Twitter they surface as major players in the online Dengvaxia debate.

Headline analysis

While social media played an important role in driving particular narratives and conspiracy theories opposed to Dengvaxia and those parts of the government involved in the implementation of the vaccine program, mainstream media also played a significant role in amplifying the mass panic surrounding the vaccine and curtailing trust in public institutions. Mainstream media published stories with sensationalist headlines supporting false claims about deaths linked to Dengvaxia, as well as stories suggesting wide-scale corruption and cover-ups. Media outlets gave constant exposure to the PAO and its ongoing campaign against the DoH, including the tenuous links that the PAO drew between the deaths of children and the Dengvaxia vaccine. The emotive videos that outlets published, including footage widely shared across social media of children allegedly dying because of the Dengvaxia vaccine, likely contributed to the public panic and loss of trust in vaccines.

Furthermore, a language analysis of 427 headlines in English published between December 2017 and September 2018 (the height of the panic around Dengvaxia) adds support to the notion that mainstream media fueled the mass panic. The common use of words and phrases such as “Dengvaxia mess,” “vaccine mess,” “dengue death,” “Sanofi,” “death,” “doh,” “garin,” and “mess” are not only emotive, but buttress what became (and to an extent still is) part of the broader misleading narrative: that the DoH, with support from former Health Secretary Janette Garin, created the “vaccine mess” which killed hundreds of children. To date, no causal relationship has been established between the deaths of Filipino children and the Dengvaxia vaccine.

As the following graph shows, some of the top two-word combinations used in headlines underline the dominant narrative that the government, specifically the DoH, bungled the vaccine program and put Filipino children’s lives at risk. Word combinations such as “Dengvaxia mess,” “Dengvaxia risk,” “vaccine mess,” and “Dengvaxia controversy” are common. The reality, of course, is much more complicated and nuanced, but this isn’t evoked in the headlines or in mainstream media coverage of the Dengvaxia vaccine program.

Fig 12: Source: First Draft

It was outside the scope of this research to analyze news and headlines published in Tagalog. The analysis of mainstream media’s portrayal thus focused exclusively on English-language stories and headlines. While English is one of the official languages of the Philippines, it is not as widely spoken as Tagalog. As a consequence, an analysis which focuses only on English-language headlines is more limited than one involving English and Tagalog. Nevertheless, because English is still an official language spoken by many Filipinos, it’s hoped that this analysis can help to elucidate the mainstream media’s role in augmenting the panic and some of the central narratives around Dengvaxia.

Facebook analysis

As this graph demonstrates, our research into engagement on Facebook posts related to Dengvaxia shows there was a spike in total interactions in December 2017, just as the Philippines halted the vaccination program amid concerns and public hysteria about the vaccine.

Fig 13: Source: First Draft using CrowdTangle data

Our analysis over the four-month period showed a total of 4,965 different Facebook Pages that posted 78,157 pieces of content with the keyword “Dengvaxia” and produced 19,845,183 total interactions. Total interactions are “the total number of likes, reactions, shares and comments on a Facebook post,” according to CrowdTangle.

Looking at just the first spike in Dengvaxia-related content shows that total interactions, between November 2017 and February 2018, neared 10 million (9,879,935).

Top 10 Facebook Pages driving total interactions around Dengvaxia between November 2017 and February 2018

Fig 14

Fig 14: Source First Draft Data from CrowdTangle

The second spike, which occurred almost two years later between January 2019 and March 2019, produced nearly 2.5 million total interactions (2,447,983). Although the amount of engagement was lower, this second spike also included misinformation and conspiracy theories.

Influential accounts such as Thinking Pinoy (run by RJ Nieto), MOCHA USON BLOG (run by actress Mocha Uson), and Sass Rogando Sasot are run by bloggers supportive of Filipino President Duterte. Thinking Pinoy has attempted to discredit various news organizations such as Rappler, and journalists such as Jes Aznar from Agence France-Presse.

Twitter analysis

Tweets with the keyword “Dengvaxia” pulled from Twitter show a similar pattern to the Facebook data acquired from CrowdTangle (see Figure 1) where an initial spike in tweets occurred between December 2017 and March 2018 during the panic that ensued after Sanofi and the WHO reissued their guidelines on the safety of the Dengvaxia vaccine. This spike was followed by another, albeit, smaller increase between January and March 2019. (See Fig. 13)

From October 2015 to July 2019, there were 92,454 tweets including the keyword “Dengvaxia,” producing a total of 163,987 retweets and 502,210 likes.

Top 10 Twitter accounts pushing out Tweets with #dengvaxia

Fig 15: Source: First Draft Data from Crowdtangle

The Twitter handles that produced the most Dengvaxia retweets are @imthinkingpinoy, which has nearly 1.5 million followers, and Francis Cruz, chief Dengvaxia whistleblower for the organization Volunteers Against Crime and Corruption (VACC). Along with Persida Acosta, Cruz has been one of the foremost critics of the Department of Health, saying there were “mafia-like operations” inside the DoH and encouraging President Duterte to take action against members of the organization. He has contributed to the ongoing scaremongering around Dengvaxia and other vaccine programs in the Philippines by supporting the idea that Dengvaxia has killed children, despite the lack of supporting evidence. Some tweets, like the one below posted by Francis Cruz, compared the Dengvaxia scandal with genocide.

Fig 16: A Tweet by Francis Cruz offering to email a “Dengvaxia tragedy” slide presentation. He also suggests viewers watch it along with the music from Schindler’s list, alluding to the idea that the dengvaxia was an intentional attempt at genocide.

Fig 17: The Twitter profile of Francis Cruz whose bio includes that the PAO forensic team has already scientifically established a strong link between child deaths and Dengvaxia, which has not been causally established.

The emergence of the #Denggate hashtag on Twitter and Facebook

The hashtag #denggate has been used by influential bloggers in the Philippines to drive the various conspiracy theories suggesting that individuals at the Department of Health made generous amounts of money through the government’s deals with Sanofi, while at the same time putting large populations of Filipino children at risk. Filipinos and Filipino children are framed as experimental guinea pigs, and the term “genocide” was thrown around liberally as part of this conspiracy. The hashtag was identified after a CrowdTangle analysis of some of the top Facebook posts with the keyword “Dengvaxia”. Once #denggate was identified, we searched both Facebook and Twitter for use of the hashtag.

The following shows all tweets that included the hashtag #denggate between December 2017 and April 2019. There have been 1,787 tweets since December 2017 with #denggate resulting in a total of 9,775 retweets and 25,186 likes. Of those total tweets, 1,357 occurred in December 2017 at the height of the confusion and panic around the Dengvaxia vaccine.

Fig 18: Source: First Draft using CrowdTangle data

Imthinkingpinoy (ThinkingPinoy on Facebook) and srsasot (Sass Rogando Sasot on Facebook) were some of the key accounts involved in fanning #denggate.

From  December 2017 to April 2019, there were 3,288 Facebook posts with #denggate, producing 2,445,490 interactions. Out of all of these posts, 2254 occurred in December 2017.

As this graph demonstrates, data around #denggate from both Twitter and Facebook shows a similar pattern, where posts and tweets spiked in December 2017 and tailed off by April 2018.  (See Fig. 18)

 

RECOMMENDATIONS

We believe that there is much for technologists, health professionals, researchers, and platforms to learn from the dengue/Dengvaxia incident in the Philippines. A study predicts that 60 per cent of the world’s population will be at risk of dengue by 2080. According to Dr. Brady, co-lead author of the study from the London School of Hygiene & Tropical Medicine: “While climate change is likely to contribute to dengue expansion, factors including population growth and increasing urbanization in tropical areas will play a much larger role in shaping who will be at risk in the future.”

The current climate of vaccine hesitancy online needs to be tempered with information that features truthful vaccination content from health authorities and trusted news sites. Gundo Weiler, the former representative of the WHO, suggested that the “big lesson” around the Dengvaxia incident was to avoid the cacophony of voices on social media and mainstream media which often communicated conflicting and politicized information about the vaccine. Part of this conflicting information came from the medical community itself in the Philippines. He also underlined how health authorities in the Philippines may not have had enough time to organize and engineer a communications strategy to mitigate the effects and impact of Sanofi’s announcement about Dengvaxia. We recommend the prominent placement of trusted news sources, as well as the promotion of accounts that feature fact-based information about the dengue fever on all platforms.

There are informed sources, particularly on Facebook, that reach thousands with content about dengue fever. However, these Pages are often eclipsed by the sharing of conspiratorial content. Professor Heidi Larson, director of the Vaccine Confidence Project at the LSHTM, said: “The Sanofi announcement was a spark that fueled the flames of underlying political ferment in the Philippines. Health authorities and immunization programs cannot solve political tensions, but trust issues and potential areas of anxiety and possible dissent must be considered in advance of a pandemic. This is especially important in an era of social media and the ability for misinformation to be spread far and wide at the touch of a button.”

We also suggest that health authorities and trusted news sites be trained in best practices for social media monitoring, debunking misinformation, and responsible reporting of hoaxes and conspiracy theories in an age of misinformation. Media outlets aired highly emotive videos showing crying mothers who claimed their children’s deaths were due to Dengvaxia. Yet, they did not provide evidence that the deaths were linked to the vaccine. Focusing on the immediate effects of a vaccination on a child is known to be an effective anti-vax strategy.

Our data shows that between December 2017 and February 2018 there was a dramatic spike in total interactions around Facebook content mentioning Dengvaxia. When this data was analyzed, it was determined that many of the posts were spurious, comprised of exaggerated claims and conspiratorial content. Creating an alert for these misinformation spikes could be an informative tool to investigate what kind of content is being posted, and provide health authorities, fact-checkers, and newsrooms with context to inform and encourage the dissemination of content that directly debunks these rumors.

Finally, in the case of the Philippines, various experts suggested that the impoverished, uneducated, and under-resourced citizens of the country suffered disproportionately from health misinformation around Dengvaxia compared to more affluent demographics. Without access to private clinics and with insufficient media literacy skills — a result of little education — these more marginalized demographics bore the undue burden of health misinformation. We thus recommend more research into how different demographics dealing with the same issue respond to such incidents, as well as how to best ensure that marginalized and isolated populations have access to good information.

Research and analysis by Jacquelyn Mason and Rory Smith

Graphics by Ali Abbas Ahmadi

Illustration by Rebecca Hendin

Stay up to date with First Draft’s work by subscribing to our newsletter and follow us on Facebook and Twitter.