The Importance of Incentives: Strategies for Excellent Surveillance

SVEA CLOSSER: Eradication programs need to have surveillance systems that cover every corner of the Earth. To achieve high-quality surveillance over this wide area, incentives can be extremely useful. We’re going to talk about the incentives used in polio eradication in Nigeria and DRC. But before we do that, let’s go back in time a bit to explore a classic example of surveillance incentives, the Smallpox Eradication Program in Bangladesh in 1975. As in our current examples of Nigeria and DRC, reaching remote parts of Bangladesh for smallpox surveillance was very difficult in the 1970s. The image on the left is a bamboo bridge being traversed by smallpox eradication team members.

The image on the right is two local Bangladeshi women working as community smallpox eradication volunteers. They were going house to house in search of smallpox cases. In the early 1970s, teams of surveillance officers traversed Bangladesh, visiting markets and schools and asking about cases of smallpox. Yet surveys showed that they were recording only a fraction of the total smallpox cases. A new surveillance strategy was needed. Faced with this problem, the Smallpox Eradication Program began publicizing cash rewards for community members who reported smallpox cases. The photo on the left shows a local Bangladeshi smallpox eradication team volunteer using a hand-held microphone to announce this reward. He’s doing this in a marketplace which drew large crowds.

If from these efforts someone reported a case, the Smallpox Eradication Program investigated it. And if they found an actual smallpox confirmed case, they paid a reward to the informant, no matter the age of the informant. Children could get these awards as well. As shown on the right, children were a target of surveillance efforts, because they went a lot of places. And they were motivated by money. So 500,000 leaflets were printed and distributed. And on the right, they’re being distributed to children. The last known case of naturally occurring smallpox, a young girl named Rahima Banu was reported by an 8-year-old who was paid her 250 Taka reward.

All the rewards were handed out in a public place so that more people would know about them. So this graph here shows the importance of these surveillance incentives very clearly. So if you look at the bottom of the graph, in 1974, the incentive was paid to the first person to report the case. And the incentive was $6.00 in 1974 money. So it was a substantial incentive in the context of rural Bangladesh. They did discover, however, that there was an issue with this. Health workers were reluctant to tell people about this reward, because the health workers wanted to be the one to get it. So they would wait until they heard about a smallpox case.

They would be the one to report it. And they would collect the reward. This meant that a lot of people in the community didn’t know about the reward and maybe weren’t turning in smallpox cases or reporting smallpox cases in a way that one would hope. So what you can see where the red arrow is, in 1975, they made a couple of changes. The first thing they did is they more than doubled the size of the reward to about $15. And the second thing they did is they gave two rewards. The first was to the member of the public who reported the case. And the second was to the health worker who confirmed it.

When they made this change, public knowledge of these rewards exploded, because the health workers now wanted everybody to know about this, because they would also get a reward. So this is an example of how important it is not just to have incentives, but to get the incentives right so that health workers and community members alike were rewarded for reporting smallpox cases. And this reward increased as smallpox cases declines. So it became very large when there were very, very few cases left. The system was very successful and, in fact, led to the elimination of smallpox from Bangladesh.

So given this model, this really successful example of incentives and surveillance in an eradication program, let’s think about why this hasn’t been done in polio eradication. So the first question to think about is, how is polio surveillance different from the smallpox system? And why aren’t direct rewards of this type widely used within polio eradication? Take a minute and see if you can think about some answers to this question. So there’s a number of ways that polio as a disease is different than smallpox that makes surveillance harder. One thing is that a lot of AFP cases that are picked up by the polio surveillance system are not polio. The vast majority of them, in fact, are not polio.

Of course, with smallpox surveillance, some of the cases picked up were not smallpox. They could be measles, other things that look like smallpox. But the proportion of cases that were smallpox was actually higher. So it becomes just too expensive to do it this way. If you think about looking at AFP cases across the entire world, the vast majority of which are not polio, it becomes rather difficult to think about giving direct rewards to people. The second question is, are there other diseases for which such a system could be effective? Now, there’s many possible correct answers here. There’s lots of cases where giving people a reward for reporting a disease is going to increase surveillance.

But generally this isn’t used outside of eradication programs, because it’s just too expensive. And that answers question 3, why isn’t this standard practice in public health? For non-eradication programs, this is too expensive and too labor intensive. Let’s take a look at polio surveillance systems. And as these charts show, they’re extremely complex and multilayered systems. So the two charts we have here are probably a little overwhelming. And the point is not that you remember each piece of them, but that you understand that these surveillance systems are very complex systems. There’s many layers and many players involved. And all of these people need to act in a timely and appropriate manner in order to successfully detect cases. And this can be challenging.

So the polio system engages health workers at several levels. So if you look at the complicated chart on the left, at the bottom is the community, the place where polio cases may be occurring. One level up from that is the health center. So health facilities are asked to report any cases of paralysis among children to a central focal point in each district. So they may or may not do this. Some of them do a very good job. Some don’t.

So what you see on the right is that the WHO also often in many areas has surveillance staff whose whole job is to go to health centers, look through the records, and see if there have been any acute flaccid paralysis cases that have shown up. So they often do this weekly. In some cases, they may do it monthly in the case of less busy health centers or maybe some private offices. They also follow up on suspected polio cases by collecting stool samples. So ideally, this all happens within the government system. But health officials being what they are everywhere in the world, they don’t always report cases. So it’s helpful to have additional WHO staff following up.

So if there’s a suspected polio case, stool samples are collected from that kid. And they’re sent to the lab to test for polio virus. So the reason we go through all of this is to think a lot about incentives or disincentives of specific stakeholders. So let’s start from the community level. The community in the case of polio doesn’t have a huge amount of incentive to report cases. However, when a child is paralyzed, this is usually a pretty traumatic incident for a family. They almost always get care in that case. So when you have a case of acute paralysis in a child, it’s almost always the case that that child is going to show up often at multiple health centers.

Certainly in the case of polio where it’s not getting better, the child is not going to show up probably just at one place. They’re going to show up at multiple places as the parent searches for care. So although there’s not a lot of incentive for the community to report cases per se, there is a lot of incentive for them to seek care for the child. And so they’re going to be interfacing with the health center, probably at multiple points. So let’s look a level up at the health center. So at the health center, there may or may not be a lot of incentive to report.

In fact, if the health center is trying to show that they’re doing a great job in eradicating polio, there may be some pressure or incentives to hide cases, which is something that is very challenging to overcome. So that’s part of the reason that you may have separate surveillance officers through the WHO who are incentivized to find cases. It’s a way of trying to deal with these conflicting incentives that people may have. So there’s some advantages and disadvantages of such a multilayered surveillance cycle. So here’s what a Nigerian health worker told us. “They have an informant in the community. Then we have focal persons in the local government, each local government.

We have focal persons in the hospital and the community too. So the informants report to the focal person. The focal person reports to the local government, who reports to the State, who reports to both the WHO and the National government– that’s the hierarchy.” So in the case of Nigeria, they’ve got a huge network that goes beyond even what I just described to include focal people in local areas who are looking for cases and then feed these cases up. So think for a moment about what the advantages and disadvantages of a system like this are. So there’s many. I’m sure you can probably think of more than me.

But one advantage is that there’s lots of opportunity for cases to get picked up and reported at multiple levels wherever parents may be seeking care for their child with polio. A disadvantage is that there’s a lot of people involved who may have conflicting incentives and a lot of ways where politics can get involved in the surveillance system. “The challenges that the polio program faces is mostly in Northern Nigeria, because most of the high-risk states are in Northern Nigeria. And this is mainly due to the Boko Haram insurgence, because as you know, if you’re unable to access a community, there’s no way you can provide vaccines to the children. And there’s no way you can also carry out surveillance activities.

And if you cannot get these indicators, then there’s no way you can be sure that there’s no wild polio virus circulating in certain parts. So the main challenge has been in the Northeast, that health workers still cannot reach those areas. And, of course, a lot of innovations have been developed to see how people like community informants from inaccessible areas and then the use of the military.” So in Nigeria, they’ve had to use a lot of other alternate methods of trying to figure out where polio cases are in areas that are inaccessible, like relying on community informants, trying to rely on the military.

And if you think about incentives, these people may not be as incentivized to report cases as a health worker might be. So in Nigeria and DRC both, conflict and insecurity are issues. And the health system has been weakened due to this insecurity. So they’ve come up with a number of strategies for conducting surveillance in hard to reach areas. For example, they’ve built new networks in camps for internally displaced families. They’ve recruited surveillance volunteers at the key points of entry and exit into the worst of the conflict zones. And in one case in Nigeria, they trained medical corps to reach conflict afflicted populations to spot signs of polio.

So let’s talk a little bit more about this idea of incentivizing community volunteers. This is a little different than the smallpox example. In smallpox, community members were incentivized for reporting a case at all. For polio, it’s a little more complicated, because community volunteers are actually working on collecting stool samples, tagging them and sending them to the lab. So this is a more professionalized and complex role for a community member. A Nigerian health official explained, “Now it is reaching every community strategy that we have in place. So they’re quite different. Now we’re taking it to the grassroots, to the community level. Unlike before, surveillance now involves an active case search in the health facility and also the community.”

So in many cases, this worked well. You got more people involved in surveillance, which is usually a good thing. But these incentives could also have perverse impacts. So some volunteers were over-tagging stool samples. So they were collecting stool samples from all kinds of kids who didn’t have polio in an attempt to receive more money for reporting more potential polio cases. This overburdened the labs and slowed down the process of identifying true cases. So getting these incentives right is very tricky. In DRC, they also tried incentivizing community volunteers. In this case, they weren’t collecting stool samples. But they were incentivized for reporting AFP cases and also for engaging in social mobilization and advocacy activities during outbreaks.

Sometimes they were given mobile phones to report AFP cases. A DRC health official said, the surveillance has been further strengthened, including community-based surveillance. They make home visits a little more. And then they also see cases and then also sensitize communities. Some people in DRC said that just like Nigeria, this worked really well in many ways. But these incentives didn’t always work exactly the way people were hoping. For example, some volunteers used the money they were given. So I want to pause here for a moment. They’re called volunteers, but they were actually paid to go door-to-door advocating for their own religious beliefs and teachings instead of raising polio awareness. So it’s not just engaging a community that’s important.

Thinking carefully about what their needs and desires are is also really important in thinking about how to engage them in a surveillance system. So the takeaway here is that community members can be great partners in surveillance, as shown by the smallpox example and also the polio examples in Nigeria and DRC. But incentives must be considered very carefully, should never assume that a community member is going to naturally want to report cases of a disease. It’s important to think both at the community level and at the health system level about what people’s incentives may be, what their motivations may be, and try to design a system that aligns with those.