Rabies control in KwaZulu-Natal, South Africa

Abstract Problem Urbanization, large dog populations and failed control efforts have contributed to continuing endemicity of dog-mediated rabies in KwaZulu-Natal province, South Africa. Approach From 2007 to 2014 we used a OneHealth approach to rabies prevention, involving both the human and animal health sectors. We implemented mass vaccination campaigns for dogs to control canine rabies, and strategies to improve rabies awareness and access to postexposure prophylaxis for people exposed to rabies. Local setting A rabies-endemic region, KwaZulu-Natal is one of the smallest and most populous South African provinces (estimated population 10 900 000). Canine rabies has persisted since its introduction in 1976, causing an average of 9.2 human rabies cases per annum in KwaZulu-Natal from 1976 to 2007, when the project started. Relevant changes Between 2007 and 2014, the numbers of dog vaccinations rose from 358 611 to 395 000 and human vaccines purchased increased form 100 046 to 156 996. Strategic dog vaccination successfully reduced rabies transmission within dog populations, reducing canine rabies cases from 473 in 2007 to 37 in 2014. Actions taken to reduce the incidence of canine rabies, increase public awareness of rabies and improve delivery of postexposure prophylaxis contributed to reaching zero human rabies cases in KwaZulu-Natal in 2014. Lessons learnt Starting small and scaling up enabled us to build strategies that fitted various local settings and to successfully apply a OneHealth approach. Important to the success of the project were employing competent, motivated staff, and providing resources, training and support for field workers.


Introduction
Rabies is a fatal zoonotic disease, causing tens of thousands of human deaths each year. 1 Most human rabies cases worldwide are caused by dog bites 2 and are preventable through canine vaccination and the provision of rabies postexposure prophylaxis to exposed persons. 3,4 Many rabies-endemic countries lack the finances and infrastructure to sustainably vaccinate dogs, conduct surveillance or provide communities with access to rabies vaccines.
Vaccinating 70% of the canine population is currently recommended to interrupt transmission of rabies. [5][6][7] However, the scale of this task can be a deterrent to setting up control programmes. Furthermore, quality baseline data on dog populations are viewed as necessary to target canine vaccination campaigns effectively and to measure vaccination coverage accurately. 8 In many rabies-endemic countries reliable data are scarce. 1 People exposed to rabies require timely prophylaxis including wound cleaning, vaccines and sometimes rabies immunoglobulins. Effective delivery of postexposure prophylaxis relies on good public awareness of rabies, and access to treatment.
Canine rabies has persisted in KwaZulu-Natal province of South Africa since its introduction in 1976. 9 Since then, urbanization, large dog populations and failed control efforts have contributed to its endemic status. We describe the KwaZulu-Natal rabies project from 2007-2014, established to eliminate human rabies through control of canine rabies and to design a programme that could be rolled out in neighbouring regions and countries.

Local setting
Located on the eastern seaboard of South Africa, KwaZulu-Natal province has an estimated human population 10 900 000. 10,11 In 2007, the year the project started, 473 cases of animal rabies were reported. The numbers of dog vaccinations done and human vaccines purchased the same year were 358 611 and 100 046, respectively. Human rabies vaccines and rabies immunoglobulins are provided free of charge in KwaZulu-Natal and the relatively small annual number of human deaths in KwaZulu-Natal (mean 9.2 over the period 1976 to 2007) was attributed to effective use, and overuse, of expensive intramuscular postexposure prophylaxis.

Approach
The project used a OneHealth approach to human rabies control, involving both the human and animal health sectors in mass vaccination of dogs, public education and awareness, and improved delivery of rabies postexposure prophylaxis. Staff members of the KwaZulu-Natal Department of Environment, Agriculture and Rural Development coordinated the programme, with funding from the Bill and Melinda Gates Foundation. The programme was managed via the World Health Organization. There was collaboration with animal welfare groups, academics, nongovernmental organizations and the human health-care sector through rabies action groups and trainings held for health professionals. We started in one village in 2007, later scaling up the programme to the entire province, and eventually neighbouring provinces and countries.
Problem Urbanization, large dog populations and failed control efforts have contributed to continuing endemicity of dog-mediated rabies in KwaZulu-Natal province, South Africa. Approach From 2007 to 2014 we used a OneHealth approach to rabies prevention, involving both the human and animal health sectors. We implemented mass vaccination campaigns for dogs to control canine rabies, and strategies to improve rabies awareness and access to postexposure prophylaxis for people exposed to rabies. Local setting A rabies-endemic region, KwaZulu-Natal is one of the smallest and most populous South African provinces (estimated population 10 900 000). Canine rabies has persisted since its introduction in 1976, causing an average of 9.2 human rabies cases per annum in KwaZulu-Natal from 1976 to 2007, when the project started. Relevant changes Between 2007 and 2014, the numbers of dog vaccinations rose from 358 611 to 395 000 and human vaccines purchased increased form 100 046 to 156 996. Strategic dog vaccination successfully reduced rabies transmission within dog populations, reducing canine rabies cases from 473 in 2007 to 37 in 2014. Actions taken to reduce the incidence of canine rabies, increase public awareness of rabies and improve delivery of postexposure prophylaxis contributed to reaching zero human rabies cases in KwaZulu-Natal in 2014.
Lessons learnt Starting small and scaling up enabled us to build strategies that fitted various local settings and to successfully apply a OneHealth approach. Important to the success of the project were employing competent, motivated staff, and providing resources, training and support for field workers.
Lessons from the field Rabies control in South Africa K LeRoux et al.
The first objective was to collect or estimate baseline data on dog populations. To understand the types of communities at risk of rabies we obtained national human census data, supported by disease data and local knowledge of the area. Additionally, we supported a study of local dog ecology to better understand canine populations, dog-human interactions and factors preventing access to dogs for vaccination. 12 This used surveys and community interviews to identify high-risk characteristics of canine populations, and dog ownership and management related to rabies endemicity in the province. We also created a database to record laboratory-confirmed human and canine rabies cases and used historic and current rabies data to inform campaign planning.
We initially measured canine rabies vaccination coverage in two villages of KwaZulu-Natal in a census to estimate the dog population. Once the effectiveness of our vaccination strategy was established, we did not expend further resources to evaluate this and thereafter we used rabies incidence in dogs and people as an indicator of the success of the project.
The second objective was to improve the systems for treating exposed people. Awareness of rabies in KwaZulu-Natal was already high, 12 but we aimed to improve the public's awareness of rabies, and to encourage them to seek treatment after a suspected exposure. We did this through rabies action groups and locally produced pamphlets, posters, radio broadcasts and newspaper advertisements directed towards medi-cal professionals and the public. To avoid inappropriate use of post exposure treatment we trained medical staff in how to perform risk assessments of exposed patients. 4 Among the different training interventions were orientations for new staff (conducted by hospitals throughout the province) on assessing and prioritizing animal-bite patients, supported by the materials generated by the project for health-care workers on managing dog bites.
The third objective was to improve the control of rabies in domestic dogs. We initially used sterilization of domestic dogs for canine population control. However, this proved slow and expensive, with little overall impact on dog population size. This is consistent with reports that rabies control can be achieved without population management. 13,14 Instead, we used existing knowledge of the local rabies epidemiology to implement targeted vaccination campaigns. For example, we vaccinated dogs in potential source areas to halt disease transmission to adjacent areas. This approach was informed by project staff who understood the local conditions and by sound surveillance systems that identified areas where rabies was more prevalent.
To conduct the vaccination campaigns, we trained and equipped technicians to use humane animal-handling equipment to catch and handle dogs. Printed and broadcast media informed the public on why, how and where to participate. We equipped field workers with vehicles and public address systems allowed staff to call the public to the nearest road to have their dogs vaccinated. By reducing travel distances for owners, we aimed to improve turnout and allow technicians to follow up on animals that were easier to handle at home. We equipped vaccinators to catch and restrain unmanageable dogs; however, free-roaming dogs that could not be caught were left alone. Finally, we created a canine rabies vaccine bank at the provincial veterinary laboratory to provide stability of vaccine supplies, lower prices for bulk orders and a base for expansion of the project.
The fourth objective was to improve the surveillance and diagnostics of rabies. We aimed to increase the submission of samples from suspect animals for laboratory diagnosis by raising awareness, and improving sample transport systems. The same transport infrastructure required to deliver vaccines to communities was used to transport samples from communities to laboratories for diagnosis.
To complement traditional surveillance methods we also used tools such as typing of viruses and sequencing, to determine transmission histories and enhance our understanding of disease causes and spread.

Relevant changes
In 2014, when the project concluded, the annual numbers of dog vaccinations performed and human vaccines purchased in KwaZulu-Natal had increased to 395 000 and 156 996 respectively (Fig. 1). The number of canine rabies cases reported in the province had fallen Start of project 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 Notes: KwaZulu-Natal province has an estimated human population 10 900 000. 11 Mass canine vaccination campaigns were implemented in the province from 2007-2014.
Rabies control in South Africa K LeRoux et al.
to 37 and there were no human rabies cases that year (Fig. 2). These figures account only for laboratory-confirmed cases; there were likely additional undiagnosed cases within the province. The increases in cases in 2012 corresponded to interrupted canine vaccination campaigns in 2011 due to an outbreak of foot and mouth disease in the province (diverting attention from dog vaccination campaigns).

Lessons learnt
We defined success of the project as reaching zero human rabies cases. However, measuring the impact of each intervention is complex. For example, improving the public's awareness of rabies involved multiple measures over a period of years, and helped to increase turnout in vaccination campaigns, demand for postexposure prophylaxis and submission of samples for surveillance.
We started activities in one area and scaled them up to build systems that fitted the local settings. This allowed rabies awareness, postexposure prophylaxis delivery, dog vaccination and surveillance to improve as the project grew. Local successes generated data, interest and investment, allowed for adaptation, and drove expansion. Effective human management, such as engaging local champions, and training, motivating and equipping field staff, was key to successfully implementing these related interventions (Box 1).
Understanding rabies epidemiology within KwaZulu-Natal allowed for targeted interventions in areas where disease transmission was highest. Our experience demonstrates that knowledge and data can be generated while interventions are being implemented; lack of data should not preclude control efforts.
We found that sustained, targeted, high-coverage dog vaccination in potential source areas consistently halted disease transmission to adjacent areas. Thus, fewer vaccinations were needed to achieve control. The dog vaccination campaigns also generated training materials and standard operating procedures, enabling neighbouring provinces and countries to implement similar campaigns.
In KwaZulu-Natal, high levels of rabies awareness, and established transport systems, including a dedicated courier service and training on transport packaging, contributed to the soundness of the surveillance system. Good surveillance systems are integral to control; however their absence should not preclude implementation of control programmes.
Improving the public's access to post exposure prophylaxis was likely a factor in reaching zero human rabies deaths in KwaZulu-Natal, as it provides protection even when a low level of disease continues to circulate in dogs. However, we found that increasing the public's awareness led to increased use of human vaccines, as dog bites continued to occur and be treated as potential exposures even when the incidence of animal rabies was reduced. Therefore, we renewed our focus on bite prevention education. Bite prevention education became a focus of the project towards the end, but was limited by lack of funds, and aims to educate communities on how to interpret dog behaviour to avoid bites. We have yet to evaluate the potential of such education to reduce rabies exposure and demand for postexposure prophylaxis.
Intradermal vaccination would reduce the cost of rabies treatment. However, it is not currently registered in South Africa or included in label recommendations on rabies vaccines. In the long term, the costs of postexposure prophylaxis could be reduced by provision of intradermal vaccines; risk assessments of animal bite patients to prevent unnecessary use of prophylaxis; and bite prevention education to prevent potential rabies exposures. ■

Box 1. Summary of main lessons learnt
• Starting small and scaling up enabled us to apply the OneHealth approach successfully and to build strategies that fitted the local setting. • Important to the success of the project were employing competent, motivated staff, and providing resources, training and support for field workers. • Our rabies vaccination campaigns generated training materials and standard operating procedures, enabling neighbouring provinces and countries to implement similar campaigns.
Lessons from the field Rabies control in South Africa K LeRoux et al.

El control de la rabia en KwaZulu-Natal, Sudáfrica
Problema La urbanización, las grandes poblaciones de perros y los esfuerzos fallidos de control han contribuido a la continuación de la transmisión de la rabia por perros en la provincia de KwaZulu-Natal, Sudáfrica.