Has rabies faded away? Where does the future of epidemic prevention go from here?
Editor's note: Rabies news has gradually faded from the media spotlight, yet the work of disease prevention must not lose focus. Chen Chen-chih has just returned from Canada with a doctorate in wildlife epidemiology. Let's hear what he has to say about how prevention work should be planned for the long term.
Profile of Chen Chen-chih
PhD in veterinary epidemiology from the University of Saskatchewan, Canada, with expertise in both veterinary medicine and ecological research; his doctoral dissertation was titled "Predicting epidemic patterns of West Nile virus in the Canadian prairie ecosystem." In Taiwan he has conducted wildlife epidemiology studies including "Spatial and temporal distribution of canine distemper infection in the Chinese ferret-badger (yellow weasel) in the Shanping area" and "Spatial and seasonal distribution of tick and intestinal helminth infections in the Chinese ferret-badger in the Shanping area." He is currently conducting postdoctoral research at the Institute of Wildlife Conservation, National Pingtung University of Science and Technology.
Q1: What can you share about Canada's experience with West Nile virus?
Chen Chen-chih (hereafter Chen): West Nile virus originally occurred in Africa, the Middle East and other regions, but after 1999 it suddenly appeared in North America, infecting birds (mainly corvids) and also being transmitted to humans by mosquitoes, causing encephalitis and posing a considerable threat. Canada therefore formed a professional West Nile disease prevention team composed of epidemiologists, entomologists, ornithologists, veterinarians, ecologists and others, conducting extensive sampling in various locations and continuous monitoring year after year, yet still unable to eradicate West Nile virus.
If we want to achieve effective control of an epidemic, in the long term we still have to build a model that allows prediction (Model).
Q2: If we want to build a model that can predict the source of a wildlife epidemic, the season of occurrence, the location, the transmission routes, and thus the disease, what background data are needed?
Chen: To build an epidemiological prediction model, you need at least monitoring data spanning multiple years, continuous data from the same area across different seasons, and a sufficiently large sample size (Sample Size). In Canada's monitoring of West Nile virus, each province's health authority originally had a group of specialists and scholars dedicated to epidemiological research, but a special task force was formed for West Nile virus as well, setting up traps widely in various regions to capture mosquitoes and birds and obtain sufficient continuous sample data. Establishing the epidemic prediction model for West Nile virus took a total of 6 years in the early stage to build the baseline data.
From the perspective of wildlife epidemiology as a profession, all statements and inferences require a long period of accumulated data before judgments can be made. Unfortunately, long-term monitoring of wildlife is something Taiwan currently lacks.
Q3: Wildlife epidemiology is more difficult to judge than human epidemiology; how should it be integrated professionally?
Chen: Besides epidemiology experts, wildlife ecology experts, evolutionary scientists, laboratory diagnosticians, veterinary scientists and environmental scientists are all very important. Therefore, epidemiological research teams abroad should be composed of these professionals, but what is currently lacking relatively is researchers in wildlife epidemiology, ecology and evolutionary biology. If wildlife epidemic judgments are approached only from the angle of human epidemiology, blind spots easily arise, so cross-disciplinary integration of opinions is needed. In this regard, academia needs to take the lead in the future. Of course, when government agencies concerned (disease control and quarantine units) recruit personnel, besides veterinary and plant pathology professionals, they also need to consider people from the professional fields mentioned above.
After all, research in wildlife epidemiology cannot be conducted the way human epidemiology can, simply by consulting questionnaires and hospital case databases. Relatively speaking, the sampling work of wildlife epidemiology is very difficult, and I hope that in the future there will be room for improvement in both staffing and funding allocation for wildlife research.
Q4: Is virus inoculation testing on laboratory animals (dogs, house shrews...) necessary? Effective?
Chen: Even if gene sequencing results show that the virus strain in Taiwan's ferret-badgers differs from the virus in dogs in other regions, forming a single lineage that may have evolved independently in Taiwan for some time and shown intra-island differentiation, this still cannot rule out the possibility that the ferret-badger virus could infect humans via dogs. To confirm this, animal experiments must be conducted: confirming whether infected dogs are capable of further transmission?
Because the current first round of animal experiments is focused mainly on dogs, under the 3R* principles of animal welfare, it is impossible to use too many live animals for experiments, but experiments with small samples still have reference value. Through the responses of the experimental subjects, we can understand the morbidity rate and further transmission capacity of dogs after infection (for example: the amount of virus in saliva), and depending on how high or low the possibility is of their infecting other species (humans), the prevention strategy will differ.
As things stand, because rabies animal experiments must be conducted in a Biosafety Level 3 laboratory, and such laboratories are scarce in Taiwan, priorities must be set for conducting them. That is to say, follow-up experiment arrangements must be adjusted according to the results of earlier animal experiments. Therefore, even if the sample size is not large, it still has certain reference value for disease prevention.
Q5: Can it be confirmed which areas in Taiwan are epidemic areas/non-epidemic areas, and are infected ferret-badgers concentrated only in low-mountain areas?
Chen: The current sampling areas are locked onto the epidemic hotspots of Nantou County and Taitung County, with Miaoli County, where no cases have yet been found, as the control group. However, if areas are judged to be non-epidemic areas based only on reporting data, that may be premature. For example: the low-mountain zone has more frequent human activity, so of course there are more cases, but that does not mean there are no infected animals in places deeper in the mountains. So-called non-epidemic areas may also be the result of insufficient sampling, leading to no infected animals being found, and one cannot be certain that no infected animals exist in that area.
The current prevention strategy is to contain the epidemic in the mountain areas, isolating potentially infected wildlife from domestic dogs, and reducing the possibility of the virus spreading to humans through domestic dogs. Comprehensive vaccination has now been carried out for dogs in Aboriginal communities in the low-mountain areas.
Q6: Can current sampling determine the infection rate in wildlife?
Chen: Interpretation of infection rates requires a certain sample size, and the detected data must incorporate the concept of confidence intervals (CI). It is not that 2 infected out of 10 means the infection rate is 20%; it could be anywhere from 5%-35%.
Of course, the larger the sample size, the more reliable the detected infection rate. However, accumulating sample size is not as easy as handing out questionnaires in a hospital; capturing wildlife requires a certain understanding of their ecological habits. The ferret-badger is the most widely distributed carnivore in Taiwan, but it was not well known to the public in the past, so at present we must rely on the field experience accumulated by past wildlife researchers to identify the ferret-badger's habitat environment, food habits and so on, before we can capture them.
Carrying out disease prevention work requires accurate epidemiological research information first. Once sufficient sample size has been accumulated and we understand the population dynamics of the reservoir host, contacts between individuals, and the factors affecting disease distribution, we can proceed to the next step of prevention work. Oral vaccine is currently the most effective tool worldwide for preventing rabies in wildlife, but before using this tool there is still much preparatory work to be done, such as confirming safety and effectiveness, and then, together with epidemiological information, applying the oral vaccine around hotspots of disease transmission, so as to achieve prevention results.
Q7: If oral vaccine helps control the epidemic, why not apply it directly in the field?
Chen: This must be explained in several parts:
First, if oral vaccine is placed directly in the field, will it be eaten by ferret-badgers? Or will it be eaten by other animals? The issue of palatability for ferret-badgers must be considered; it is not that once you put it out, it works;
Second, whether a target species like the ferret-badger can develop sufficient immunity after eating the oral vaccine;
Third, because oral vaccine may affect other species or the environment, if it is eaten by other non-target species it may lead to a risk of infection and disease, since some oral vaccines are genetically recombinant vaccines made by using vaccinia virus as a gene vector with the rabies glycoprotein gene inserted. But vaccinia virus carries a risk of causing human infection, so a careful risk assessment must be conducted before application.
The current prevention measures are still at the stage of building baseline data. Only after confirming basic factors such as the transmissibility of rabies virus, the infection rate in wild populations and pathogenicity can the necessity of releasing oral vaccine be assessed. According to the recommendations of American experts, a complete assessment requires at least 1 to 2 years.
Q8: If dog and other animal experiments confirm that transmission does not occur via dogs, then in what direction should prevention work proceed?
Chen: If animal experiments confirm that dogs do not transmit Taiwan's rabies virus strain, then we can focus specifically on preventing rabies in wildlife populations. In the past, we all thought about prevention strategy in a human-centered way, but neglected disease monitoring of wildlife. In fact, this kind of zoonotic infectious disease existing in wildlife populations has a great impact on humans, domestic animals and wildlife, and Taiwan is very short of manpower for wildlife disease surveillance.
At present, gene sequences indicate that rabies may never have disappeared from Taiwan, but we have kept failing to discover the fact that rabies exists in Taiwan, and we also do not know why it suddenly broke out this time. That is precisely because of the lack of long-term monitoring of wildlife. Therefore, manpower and work for long-term monitoring are extremely important.
Q9: Can rabies and canine distemper be regarded as selection mechanisms of natural evolution? If human intervention is needed, to what extent should it intervene?
Chen: Canine distemper, like rabies, is an infectious disease that is very harmful to wildlife, and in the United States it once caused the extinction of the black-footed ferret (Mustela nigripes) in the wild. But because canine distemper does not infect humans, it is not a notifiable infectious disease under law in Taiwan, so much less attention has been paid to it.
Research by the Pingtung University of Science and Technology team in the past showed that canine distemper in Taiwan has a higher chance of appearing in secondary forests near human settlements, while in primary forests it occurs in autumn and winter and disappears in summer. The main source of the disease in secondary forests is domestic dogs, and it spreads throughout the year. Monitoring results show that the canine distemper vaccination rate of domestic dogs near Aboriginal communities is low, resulting in high infection rates (80%) and high transmissibility (spread through saliva, eye secretions, excreta and so on), making them a key reservoir host that infects ferret-badgers and yellow weasels and causes a very high mortality rate in these two animals.
Canine distemper in fact causes great harm to wildlife. It is very good that this rabies outbreak has made everyone begin to pay attention to wildlife epidemiological monitoring. Besides the impact on humans, attention must also be paid to the impact on wildlife populations.
Q10: In the Endemic Species Research Center's monitoring over the past year, ferret-badgers were found frequently road killed (Road Killed). Could this be related to rabies? Can other factors also cause road deaths of wildlife?
Chen: Rabies and canine distemper both cause encephalitis, leaving animals mentally confused and reducing their sensitivity to the environment, thereby increasing the probability of road deaths. North America has a disease widespread in cervids - chronic wasting disease (abbreviated as CWD) - which causes the central nervous system of infected deer to gradually degenerate, resulting in slow movement, weakness and reduced responsiveness to the environment, making them easy to prey on or be hit by cars on the road.
Of course, pesticides, accumulation of heavy metals and road development all increase the chance of animal road deaths. Therefore, Canadian regulations explicitly prohibit the use of lead bullets when hunting, as well as the use of poison baits, to prevent the accumulation of lead in the environment and the accumulation of toxins in the food chain.
* Animal welfare was first promoted for laboratory animals. In 1947, the Universities Federation for Animal Welfare in London published the laboratory animal handbook, laying the foundation for laboratory animal welfare; and in the 1950s it proposed the 3R principles of animal experimentation: Replacement, Reduction and Refinement. http://pansci.tw/archives/18379