Optimal control strategies for rodent-borne zoonoses in Brazilian slum settlements

One billion of the world's population lives in slum settlements, comprising 78% of the urban populations from the world's poorest countries. Residents suffer an increased burden of a wide range of diseases. Many of these health problems involve zoonotic disease - infections circulating in wildlife or domesticated 'reservoir' hosts that can be transmitted to humans, where they can cause serious illness or mortality. Most of the infectious diseases that have emerged in humans recently are zoonotic (e.g. HIV-AIDS, Ebola). Most circulate naturally in wildlife, often rodent reservoirs. Hence, learning to control the spread of infection from rodent reservoirs to humans in urban slums, particularly in poor countries, is a challenge that must be addressed.
Leptospirosis is perhaps the most widespread zoonotic disease. Hosts are infected with bacteria, leptospires, which infect the kidneys of reservoir hosts and are shed with the urine. This is the ultimate source of infection for humans, though the main mode of transmission is contact with contaminated water or soil. Leptospirosis is a systemic disease in humans. Manifestations range from relatively mild to severe life-threatening forms for which case fatality may be >50%. Its burden falls mainly on the world's poorest people, traditionally rural-based subsistence farmers, but recently shifting from a rural to an urban setting. Our study will focus on leptospirosis, which is important in its own right, but the project is directly relevant to other zoonoses in slum settings, and to the health of these communities more generally. The work will be carried out in the urban slums (favelas) of the city of Salvador, Brazil. More than 12,000 cases of leptospirosis are reported in Brazil during urban epidemics each year of which around 12% are fatal.
Realistically, the threat from leptospirosis and similar diseases in impoverished urban slums will not be countered by technological or sophisticated measures. Rather, a global health priority, and an overarching aim of our program, is to develop the means of enabling agencies working locally to minimise the threat that these reservoirs of infection pose, using methodologies that can be applied effectively, cheaply, and without the need for a sophisticated infrastructure. To this end, we will determine the consequences of various management practices (controlling either rats or the environment) so that optimal interventions can be identified that reduce total rat population size, and hence the quantity of shed leptospires and the risk to humans, in the most cost-effective way.
We are in an arguably unique position to do this. We have a very broad multi-disciplinary team from medics to social scientists, combining expertise from the UK, Brazil and the USA, and we have strong links to local public health agencies. Hence, through our project, we will determine the efficacy (evidenced by a range of metrics) of a range of control strategies, guided by, and in collaboration with, local agencies.
The overall structure of our study will be a comparison of the range of sanitation-control measures currently being employed in different favelas across the municipality of Salvador. Following consultations locally, we have identified three categories that have been operational for at least the last twelve months, and that provide us with sufficient sites to support statistically valid comparisons. At each site, the following data will be collected: (i) rat abundance; (ii) the proportion of the rat populations carrying leptospires and other key pathogens (iii) concentrations of pathogenic leptospires in the environment; (iv) the proportion of the human population that have been infected with leptospires and these other key pathogens; (v) severe cases of leptospirosis reported in humans over the last 20 years; and (vi) attitudes of local residents to methods of sanitation control, especially in relation to effects on disease risk.

It is widely accepted that controlling the spread of infection from rodent reservoirs to humans in urban slums, particularly in poor countries, is a challenge that must be addressed. We focus on leptospirosis in the urban slums of the city of Salvador, Brazil, which is important in its own right, but our project is directly relevant to other zoonoses and to the health of these communities more generally.
Our overarching aim is to develop the means of enabling agencies working locally to minimise the threat that these reservoirs pose, using methodologies that can be applied effectively, cheaply, and without the need for a sophisticated infrastructure. We are in an arguably unique position to do this with an international multi-disciplinary team from medics to social scientists, and with strong links to local public health agencies.
The overall structure of our study will be a cross-sectional comparison of three sanitation-control measures currently being employed. These three have been operational for long enough, and provide us with sufficient sites (three of each) to support statistically valid comparisons. At each site, we will collect data on (i) rat abundance; (ii) the prevalence in rats of leptospires and other key pathogens (iii) concentrations of pathogenic leptospires in the environment; (iv) seroprevalence in humans for leptospires and these other key pathogens; (v) severe cases of leptospirosis reported in humans over the last 20 years; and (vi) attitudes of local residents to methods of sanitation control, especially in relation to effects on disease risk.
These data will be combined, first, simply to compare the efficacies of current practices. But we will additionally use optimal control modelling, combined with the data on human attitudes, to explore the cost-effectivenesses of the practices, and whether new or combinations of strategies could further improve cost-effectiveness.

In the broadest sense, the ultimate beneficiaries of this research will be the urban poor in low and middle income countries (LMICs), suffering a grossly disproportionate burden of serious infectious disease, and also those responsible for improving the health of these people. Within the narrower confines of the current project, the beneficiaries will be an exemplar group: those suffering a heavy burden of serious forms of leptospirosis (case fatality c12%) in the favelas of the Brazilian city of Salvador, and the local agencies responsible for leptospirosis control.

At present, one billion of the world's population lives in slum settlements, comprising 78% of the urban populations from the world's least developed countries. Residents suffer an increased burden of a wide range of diseases because of overcrowding and inadequate access to sanitation and safe drinking water. Many of these health problems involve zoonotic disease - infections circulating in wildlife or domesticated 'reservoir' hosts, where they are often asymptomatic, that can be transmitted to humans where they can cause serious morbidity or mortality. Rodents, most often rats, are by common consent the most important group of zoonotic wildlife reservoirs, especially in urban settings. Leptospirosis in the city of Salvador, Brazil, is important in its own right, but the project is directly relevant to other zoonoses in slum settings in LMICs, and to the health of these communities more generally.

Realistically, the threat from leptospirosis and similar diseases in urban slums in LMICs will not be countered wholly, or even predominantly, by technological or sophisticated mean. Rather, local residents and local agencies will benefit most from methods and approaches that can be applied effectively, cheaply, and without the need for a sophisticated infrastructure. Utilizing our strong links to local public health agencies, we will evaluate a range of control measures that are currently being applied, and therefore provide these agencies, for the first time, with the means to judge their efficacies. We will also work with them, using our optimal control modeling approach, to estimate effective costings for each control strategy, allowing them, again for the first time, to develop novel, more cost-effective strategies. These approaches, and decisions, can be rolled out to other cities and a broader range of settings with a corresponding widening of impact.

Local residents will also benefit from our work. The success of any control measure depends, especially for its long-term sustainability, on its acceptability (or even attractiveness) within the local community. Working with local residents, we will adopt a participatory action research approach, which prioritises the views of local residents and allows them to take a proactive role in developing feasible and acceptable future interventions. Hence, they will benefit from interventions that are not only theoretically but practically most effective, but also benefit, we hope, in a deeper sense of taking ownership of the means of improving their health and the health of their communities.