Rabies is a viral disease that may affect the central nervous system of any species, but only circulates in mammals [1]. Rabies virus is mainly passed from animal to animal or animal to human through bites or scratches. In addition the virus can also be transmitted by the contamination of wounds. Under very exceptional circumstances, the virus can cross mucous membranes when the patient inhales aerosol [2, 3]. Rabies epizootics may be divided into two interrelated cycles, urban and sylvatic. The red fox (Vupes vulpes) is one of major vectors of the disease and is it reservoir for sylvatic rabies in Eurasia and in parts of America, but it is not the most frequent risk for transmitting rabies virus directly to humans [3, 4]. The more serious rabies risk to human is imposed by urban rabies. The domestic dog plays a principal role as a reservoir and transmitter of urban rabies to humans in China [5]. Humans are also at risk from affected domestic animals or pets such as cattle and cats at large, or wild animals such as the raccoon dog in Eurasia and different terrestrial or flying mammals in the New World [4, 6, 7]. Moreover, direct human-to-human transmission has been observed [8]. There is no effective treatment after the onset of the associated clinical symptoms. Therefore, the currently recommended intervention strategy is to remove and neutralize the infectious virus before it enters the nervous system [2].

According to the official World Health Organization (WHO) data [9–11], more than 2.5 billion people are at risk in over 100 countries reporting the disease. Rabies has the tenth highest mortality of all infectious diseases worldwide. There are still about 50000 to 60000 human deaths annually although effective vaccines for post-exposure treatment are available [12]. Developing countries account for almost all of the reported human deaths, and most affected are the tropical countries or regions in Africa, Asia, South America and Oceania. During the period 1993–2002, the countries of the Americas reported a decrease of 82% in the number of human cases, with cases plummeting from 216 in 1993 (mortality rate of 0.03 per 100000 inhabitants) to 39 in 2002 (mortality rate of less than 0.01 per 100000 inhabitants) [13]. Rabies is considered as a source of economic loss and, above all, hampers the movement of animals between different countries or regions, which has serious implications for the 'open market' since some countries are currently rabies free and wish to maintain their disease-free status [3].

Rabies is a major public-health problem in most of the developing world [9, 14–16]. Prophylactic measures taken in the past, such as destroying foxes and reducing dog populations, did not prevent the spread of the rabies, although recently developed genetically modified rabies virus vaccines provide an effective method of prevention of rabies virus infection in dogs, foxes and raccoons [17–19]. During recent years, most research into the control of rabies has concentrated on the development of post-exposure prophylaxis (PEP) of rabies [3]. The use of human rabies immunoglobulin (HRIG) and of equine rabies immunoglobulin (ERIG) has saved the lives of countless patients who would have died if treated with vaccine alone. However, both products are often in short supply worldwide and virtually unaffordable in developing countries [20]. Therefore, the high demand for PEP in Africa and Asia exerts a substantial economic burden, not only as a result of the high costs of human vaccine and rabies immunoglobulin (RIG) products, but also because of considerable indirect (patient) costs associated with travel and income loss for PEP [21]. Additional economic losses relate to livestock deaths, which, although poorly quantified, may be significant, with an estimated annual incidence of 5 deaths per 100000 cattle, costing US$12.3 million annually in Africa and Asia. The total (direct and indirect) cost of PEP accounts for 5.8% of annual per capital gross national income in Africa (US$40 per treatment) and 3.9% (US$49 per treatment) in Asia [22].

In China, over a 55-year period between 1950 and 2004, 108412 human rabies cases were reported with three major epidemics occurring during this period [23, 24]. The first epidemic outbreak occurred in the mid-1950s when cases rose to a peak of about 2000 annually. After a decline in the 1960s, the number of cases again started to increase in the early 1970s reaching a peak in 1982, and then remained at the level of 5000–6000 cases per year until the end of the decade [25]. Therefore, one of the purposes of this study was to conduct a comprehensive analysis of the rabies situation in the country using all of the official data to characterize the current epidemiological trends of rabies in China from 1990 to 2007. In order to define better recommendations for improving the PEP schedules delivered to patients, we also analysed the reasons for the post-exposure treatment failures (or the absence of PEP), based on the medical records of anti-rabies treatment of injuries or related incidents for 244 rabies patients, ascertained in Guangdong province of China in the years of 2003 and 2004.

In China, human rabies was largely under control in the period 1990–1996, owing to nationwide rabies vaccination programmes. Since the vast majority of cases were a result of canine rabies, an extensive dog vaccination programme was initiated in 1987 [27]. From 1990 to 1996, canine rabies decreased by 95.5%, while the GDP (gross domestic product) increased considerably in the same period. During the period 1996–1999, the yearly trend for human rabies was relatively stable, but the number of human rabies cases jumped dramatically since the start of the new millennium. Overall, our data implicate that while China was largely free of human rabies in the final years of the last millennium, the recent increase in incidence has been high enough to trigger a warning sign for control and prevention.

However, the data ascertained at various CDCs may not well capture the real epidemiological scenarios for human rabies in China. For example, since the SARS (severe acute respiratory syndrome) epidemic of 2003, the Chinese government has set up a systematic and nationwide surveillance network for zoonotic diseases. Increased surveillance together with increased dog or other pet populations may be the principal factors explaining the increasing number of cases of human rabies reported in China in the recent years [27]. The same issues for obtaining unbiased estimates for the epidemiological indices of rabies (incidence or mortality rates) were noted by Torrence et al. [28] in similar studies, and by Brazuna et al. [29] in a study of Brazilians. It should be noted that this bias is not unique to the data ascertained at various CDCs, but is instead common to many medical experiments (for example, hospital-based studies) where random sampling is not a feasible (often less-efficient) strategy for data collection. However, there is a list of factors influencing precise and accurate estimation of human rabies parameters. These include the following factors: (1) many people, especially in remote areas, do not have easy access to a public health service; (2) there was a dearth of knowledge about rabies in the general public and among health workers; (3) there was insufficient diagnostic capability and facilities in the country so that some case reports may not be able to be sent to the surveillance database at the Ministry of Health of China. Despite these limitations, the results clearly demonstrated that rabies constituted a real public problem in China and its control should be a top priority.

Our data also indicate that most incidents occurred in the southwestern and southern territories of China, and most frequently in the highly populated areas. For example, in 2005, rabies was routinely identified in Guizhou (481 cases), Guangxi (480 cases), Hunan (379 cases) and Guangdong (306 cases) provinces. The four rabies-endemic provinces lacked strictly enforced measures to eliminate dog rabies or an ample supply of modern cell culture rabies vaccines for humans. It was also interesting to note that most of patients were young in the range of 0–20 years old and the major victims of rabies were children less than 16 years old, perhaps as they are bitten by dogs more frequently than adults. When attacked by dogs, the lesions were often on the child's head and neck, thus, not surprisingly, this lesion site was associated with the highest risk for developing rabies [21, 30, 31].

In our study of the 244 human rabies cases in Guangdong province, 67.2% of the patients did not seek medical services or did not receive any PEP. The time lag between the incidents and the presentation of patients for anti-rabies assessment ranged from 0 to 3 days in most cases. Up to 62.5% of the patients did not receive proper treatment of the wounds, 92.5% did not receive adequate post-exposure vaccination and 91.25% did not receive any anti-rabies immunoglobulin. These results suggest that the population investigated may not be aware of the risks of rabies transmission, as revealed previously [23]. This fact could be a significant issue for public health based on the large number of failures for human PEP that have occurred recently in China. During the study, we found that education/information regarding rabies can only be obtained from the public boards at municipal or district CDCs in Guangdong province. We found few such public information boards or web pages at police departments (stations), community hospitals or offices, in villages and so on. In December 2007, we conducted a survey among 270 students of preventive medicine at our institute. We observed that the students had a basic knowledge of rabies, but had some misunderstandings. Among the students, 92.2% could answer questions regarding PEP regimes correctly. However, only 47.4% knew that washing the wounds can remove the virus residue on the lesions, and the percentage of students answering this correctly for students in grades 2–5 was 42.6%, 32.0%, 60.0% and 70.4%, respectively. Only 58.9% of the students knew the pre-exposure administration regime for rabies vaccine. However, 71.5% of the students knew the post-exposure administration regime for rabies vaccine and 57.8% knew the immunization sites for rabies vaccine.

In rabies-infested developing countries, modern cell culture vaccines are too costly for the poorly developed remote regions, so dangerous NTVs are still used [32–34]. Over the last 4 years in China, of the more than 2000 people yearly who died of rabies, only about one-third received rabies vaccinations. This figure was even lower (7.5%) in Guangdong province, as estimated in this study. In most cases of vaccine failures, patients contracted the disease before the full PEP regimen was completed [24]. In addition, there was a critical shortage of human and purified equine rabies immunoglobulin in these regions, which are essential in the treatment of severe exposure. Although the costs of modern vaccines are decreasing, the current price of a full-dose intramuscular vaccine treatment is somewhat beyond what an average family in developing countries can afford [35, 36]. For example, the average annual income per capita in Guangdong province is US$3000 and the rabies vaccine costs US$12.5–45.0, taking up 0.42–1.51% of the average annual income per capita. This figure would be much higher we consider the fact that about 60–70% cases were from poor rural areas or habos. Furthermore, the supply of modern and safe vaccines for many provinces are grossly inadequate, whereas the demand for affordable and safe human post-exposure treatment is increasing in these provincial regions [23, 24]. Consequently, only anti-rabies vaccines and human immunoglobulin are available, and provide feasible solutions for efficiently controlling rabies in the majority of municipalities of China or other developing countries [16, 37–39].

Finally, a mass vaccination dog campaign for the control and elimination of rabies transmitted by dogs is very important. China has not implemented enforced immunization for dogs. Rabies vaccine was provided at the cost of the owners, and the cost was expensive, from tens to hundreds of Chinese Yuan. Dogs were registered at local police departments, while animal rabies vaccines were administrated by veterinarians, thus there was a lack of good communication or effective strategies for rabies control in dogs. It appears that the increase in canine rabies, increase in the dog population and decrease in the vaccine coverage of dogs may have contributed to the apparent upwards trend in human rabies in recent years in China. Based on Chinese CDCs' surveillance of dog rabies in some typical affected areas, the positive rate was 3.9% (4/102, Guangxi province) in 1999, 9.1% (76/838, Hunan, Henan, Guangxi, Guizhou and Jiangsu provinces) in 2004 and 5.9% (5/85, Guizhou province) in 2005. This pattern is fairly consistent with human rabies. Nevertheless, the methods of control undertaken by Latin American countries might provide promising ways for China to develop a more effective programme for controlling human rabies transmitted by dogs. With the support from the Pan American Health Organization, several measures such as the decentralization of health units with PEP available, dog vaccination coverage and dog rabies surveillance have been successful in achieving the goal of eliminating human rabies transmitted by dogs [40].

This large-scale epidemiological study has demonstrated that human rabies was largely under control in China between the years of 1990 and 1996, via the national rabies vaccination programmes. However, the number of reported rabies cases has started to increase since then. From 2001, this figure increased significantly to a new peak that was reached in 2004, where it has remained. Based on the investigation of 244 human rabies cases collected in Guangdong province in 2003 and 2004, 67.2% of patients did not seek medical services or did not receive any PEP. Further analysis of the post-exposure treatment failures for 80 rabies patients who received any type of PEP indicated that the majority of patients, if not all, did not receive proper and timely treatments on the wounds or post-exposure vaccination. The study implicated that: (1) the incidence of human rabies, which China was largely free from in the final years of the last millennium, is now increasing which should be a warning sign for control and prevention; (2) there is a need to improve the current rabies control programme in order to reduce non-compliance rates and to decrease the occurrence of flaws in the surveillance programme and in the provision of health care. Given the findings of the study, we view the implementation of the following measures as appropriate [9–12, 41, 42]: (a) continuing supervision of the current human rabies control programme, in order to reduce both non-compliance rates and the occurrence of flaws in health-care provision; (b) improving the interactions between professionals from the divisions of the municipal health-care network and teams of national programmes; (c) increase rabies awareness among Chinese health authorities and policymakers, improve the training of general practitioners and health care workers, and educate school children and the general public, which is crucial for effective rabies control; and (d) urban planning and development should take ecosystem preservation into account, in an attempt to balance the interaction between humans and animals.