Vaccine scares have emerged as a major challenge to global efforts to eliminate preventable diseases, with rumours and conspiracy theories proliferating faster than health authorities can respond to them. Now researchers, led by Heidi Larson of the London School of Hygiene and Tropical Medicine, are developing a tool to identify the first signs of these negative reports.
Vaccine scares have popped up in both the richest parts of the world and the poorest. Over a decade ago, suggestions in the UK that the combined MMR (measles, mumps and rubella) vaccine could trigger autism led to a dramatic drop in the number of parents having their children vaccinated. Wales, which had one of the lowest vaccination rates, is now in the grip of a major measles outbreak, with young teenagers - the generation that was not protected - particularly affected.
Northern Nigeria saw rumours that the polio vaccine was part of a Western conspiracy to sterilize Muslims, preventing polio’s eradication in the country and leading to the disease’s reappearance in surrounding countries where it had already been eliminated.
“Bad news stories damage vaccination programmes as much as biological hazards, and these stories evolve over minutes or hours, needing immediate action,” said University of Toronto public health specialists Natasha Crowcroft and Kwame McKenzie, in a comment published this week alongside Larson’s paper in the medical journal The Lancet. “By the time a detailed scientific analysis of a vaccine safety issue is completed, the story is no longer newsworthy.”
Crowcroft and McKenzie point out that modern communications, especially the internet, can exacerbate vaccine scares. But Larson’s Vaccine Confidence Project set out to establish whether the internet could also provide the tools to fight misinformation.
Rumour surveillance
Larson’s team set up a media surveillance system covering 144 countries, looking at online articles, blogs and reports about vaccines and vaccine-preventable diseases.
The first stage of the process was automated, using the HealthMap data collection system, which searched for terms such as “vaccine”, “rotavirus” or “measles”. The accumulated material was inspected by real people, who assessed whether it positively or negatively portrayed vaccination, and whether it should be flagged as a cause for concern.
When one report appeared on multiple websites, all copies were counted, “recognizing the fact that replicated reports show the spread of information,” Larson’s paper says.
The Vaccine Confidence Project initially ran from April 2010 to April 2011. At the end of the year, they could see that the system had worked - clusters of reports expressing concern about vaccination correlated with real-world events. Of the reports analysed, 69 percent were assessed as favourable to vaccination and 31 percent as hostile.
“We picked up concerns we already knew were there, but more than that,” Larson told IRIN. “For instance, we saw activity around a narcolepsy/H1N1 vaccine link, and we were picking up early discussions suggesting this might be an issue before the final confirmation (in Finland) that there was indeed a link.
“And in Pakistan, where we were following issues around polio acceptance, we started picking up political tensions and concerns among lady health workers. We certainly didn’t predict the killing of polio workers, but we had seen the tensions growing.”
Waves of information
There are questions about whether internet surveillance, using search terms in English, can spot emerging concerns in rural societies where internet penetration is low and public debate occurs in local languages. Could this kind of surveillance, for instance, have picked up the early signs of polio vaccine rejection in Hausa-speaking northern Nigeria?
Larson, who has worked in that area on behalf of the UN Children’s Fund, says she thinks it would have.
“It was emerging in the local media a bit, and then reports started to circulate on the BBC Hausa service. And since Nigeria has English as an official language, they were soon circulating in English as well. A former Nigerian minister of health, Nike Grange, is on our advisory board, and she says that if they had had a system like this at the time, and had understood the full impact of the rumours they heard, they would have acted much sooner,” Larson said.
“And the world has changed a lot in the last decade. What we are seeing is that you don’t have to have a computer in every household. People hear something on the radio, they tell their neighbour, they tweet it, and there are waves of information. We hadn’t anticipated how ubiquitous cellphones and smartphones were going to be, and that makes this work even more relevant.”
eb/rz
Vaccine scares have popped up in both the richest parts of the world and the poorest. Over a decade ago, suggestions in the UK that the combined MMR (measles, mumps and rubella) vaccine could trigger autism led to a dramatic drop in the number of parents having their children vaccinated. Wales, which had one of the lowest vaccination rates, is now in the grip of a major measles outbreak, with young teenagers - the generation that was not protected - particularly affected.
Northern Nigeria saw rumours that the polio vaccine was part of a Western conspiracy to sterilize Muslims, preventing polio’s eradication in the country and leading to the disease’s reappearance in surrounding countries where it had already been eliminated.
“Bad news stories damage vaccination programmes as much as biological hazards, and these stories evolve over minutes or hours, needing immediate action,” said University of Toronto public health specialists Natasha Crowcroft and Kwame McKenzie, in a comment published this week alongside Larson’s paper in the medical journal The Lancet. “By the time a detailed scientific analysis of a vaccine safety issue is completed, the story is no longer newsworthy.”
Crowcroft and McKenzie point out that modern communications, especially the internet, can exacerbate vaccine scares. But Larson’s Vaccine Confidence Project set out to establish whether the internet could also provide the tools to fight misinformation.
Rumour surveillance
Larson’s team set up a media surveillance system covering 144 countries, looking at online articles, blogs and reports about vaccines and vaccine-preventable diseases.
The first stage of the process was automated, using the HealthMap data collection system, which searched for terms such as “vaccine”, “rotavirus” or “measles”. The accumulated material was inspected by real people, who assessed whether it positively or negatively portrayed vaccination, and whether it should be flagged as a cause for concern.
When one report appeared on multiple websites, all copies were counted, “recognizing the fact that replicated reports show the spread of information,” Larson’s paper says.
"Bad news stories damage vaccination programmes as much as biological hazards, and these stories evolve over minutes or hours, needing immediate action"
Although it was a worldwide survey, the researchers paid particular attention to five countries - China, Finland, France, Nigeria and Pakistan - that had seen issues over public confidence in vaccines. They also mapped reports about the human papilloma virus (HPV) vaccine in India, where trial HPV vaccination projects had been suspended in two states.The Vaccine Confidence Project initially ran from April 2010 to April 2011. At the end of the year, they could see that the system had worked - clusters of reports expressing concern about vaccination correlated with real-world events. Of the reports analysed, 69 percent were assessed as favourable to vaccination and 31 percent as hostile.
“We picked up concerns we already knew were there, but more than that,” Larson told IRIN. “For instance, we saw activity around a narcolepsy/H1N1 vaccine link, and we were picking up early discussions suggesting this might be an issue before the final confirmation (in Finland) that there was indeed a link.
“And in Pakistan, where we were following issues around polio acceptance, we started picking up political tensions and concerns among lady health workers. We certainly didn’t predict the killing of polio workers, but we had seen the tensions growing.”
Waves of information
There are questions about whether internet surveillance, using search terms in English, can spot emerging concerns in rural societies where internet penetration is low and public debate occurs in local languages. Could this kind of surveillance, for instance, have picked up the early signs of polio vaccine rejection in Hausa-speaking northern Nigeria?
Larson, who has worked in that area on behalf of the UN Children’s Fund, says she thinks it would have.
“It was emerging in the local media a bit, and then reports started to circulate on the BBC Hausa service. And since Nigeria has English as an official language, they were soon circulating in English as well. A former Nigerian minister of health, Nike Grange, is on our advisory board, and she says that if they had had a system like this at the time, and had understood the full impact of the rumours they heard, they would have acted much sooner,” Larson said.
“And the world has changed a lot in the last decade. What we are seeing is that you don’t have to have a computer in every household. People hear something on the radio, they tell their neighbour, they tweet it, and there are waves of information. We hadn’t anticipated how ubiquitous cellphones and smartphones were going to be, and that makes this work even more relevant.”
eb/rz
This article was produced by IRIN News while it was part of the United Nations Office for the Coordination of Humanitarian Affairs. Please send queries on copyright or liability to the UN. For more information: https://shop.un.org/rights-permissions
Share this article
