At the end of March 2009, an outbreak of novel influenza A (H1N1) (here after called A/H1N1) infection occurred in Mexico, followed by ongoing spread to all over the world in a short period [1]. On June 11 2009, the World Health Organization raised its pandemic alert level to the highest level, phase 6 [2], meaning that the A/H1N1 flu had spread in more than two continents and reached pandemic proportions. As of June 13, 2010, it had caused over 18,172 deaths in more than 214 countries and overseas territories or communities [3]. Most illness, especially the severe illness and deaths, had occurred among healthy young adults, which was markedly different from the disease pattern seen during epidemics of seasonal influenza [4, 5].

China is highly susceptible to A/H1N1 because of its huge population and high residential density, besides the high infectiousness of this novel influenza virus. After the first imported case reported on May 11, 2009, the confirmed cases were reported in various provinces of China [6]. By the late of October 2009, A/H1N1 cases had increased dramatically, with 44,981 cases and 6 deaths confirmed at the end of October 2009. The A/H1N1 infection rate peaked in November 2009, when approximately 1500 new cases of A/H1N1 were being confirmed each day. By the end of this month, a total of 92,904 cases and 200 deaths had resulted from A/H1N1-related causes [7]. The Chinese government has taken a series of preventive measures according to WHO guidelines, including the promotion of public knowledge about flu through mass media, patient isolation, quarantine of close contact person, and free vaccinations to population at high risk (e.g. young children, healthcare workers, and people with chronic disease) [8]. However, there were few public reports on the assessment of the effect of these policies and the level of knowledge, attitude and practice (KAP) associating with A/H1N1 among general population.

It is well-known that confused comprehension and negative attitude towards the emerging communicable disease may lead to unnecessary worry and chaos, even excessive panic which would aggravate the disease epidemic [9]. For instance, during SARS epidemic from 2002 to 2004, the misconceptions and the excessive panic of Chinese public to SARS led the public resistant to comply with the suggested preventive measures such as avoiding public transportation, going to hospital when they were sick, which contributed to the rapid spread of SARS and resulted in a more serious epidemic situation, making China one of the worst affected countries with over 5327 cases and 439 deaths [10, 11]. In addition, the panic of infectious disease outbreak could cause huge economic loss, for example the economic loss of SARS has been estimated at $30-$100 billion in US, though less than 10,000 persons were infected [12]. SARS experience has demonstrated the importance of monitoring the public perception in disease epidemic control, which may affect the compliance of community to the precautionary strategies. Understanding related factors affecting people to undertake precautionary behavior may also help decision-makers take appropriate measures to promote individual or community health. Therefore, it is important to monitor and analyze the public response to the emerging disease.

To investigate community responses to A/H1N1 in China, we conducted this telephone survey to describe the knowledge, attitudes and practices of A/H1N1 among general population in China and put forward policy recommendations to government in case of future similar conditions.

Novel A/H1N1 has caused pandemic in this century. It is important to encourage the public to adopt precautionary behaviors, which is based on the correct knowledge of the epidemic and appropriate response among residents. Many studies have examined the various levels of KAP about infectious disease outbreaks, such as SARS, avian influenza [13–15]. Some studies have been reported specifically on community responses to A/H1N1 in Australia and Europe [16, 17]. But through literature search, we haven't found any public reports on KAP regarding A/H1N1 among Chinese population until now. Therefore, we conducted this large population-based survey (10669 respondents) to investigate community responses to A/H1N1 and to provide baseline data to government for preventive measures in case of future outbreaks.

Unless people have basic knowledge about the modes of transmission, they respond appropriately during an outbreak [16]. It has been proved that influenza is transmitted through person to person via respiratory secretions [18]. Most residents in our survey recognized that the risk of getting infected would increase when an infected person coughed or sneezed in close distance. This may be due to the previous experience of SARS and avian flu. Multivariate analysis results showed that workers and farmers with lower education level were less likely to have this knowledge, which indicated that the contents and forms of propaganda should be more understandable and acceptable. A large proportion of residents in our survey overlooked the indirect hand contact and hand-shaking transmission route and about one third of public misconceived that A/H1N1 was food borne, which was associated with the previous knowledge of avian flu and the new A/H1N1 flu in the general population. The confusion with avian flu might mislead some residents to believe that the A/H1N1 virus is fatal and cause public panic [19]. Therefore, it is important for the government and health authorities to provide continuously updated information of the emerging disease through televisions, newspapers, radios, and Internet.

There are regional differences in the perception of A/H1N1. For example, the public in Hong Kong did not perceive a high likelihood of having a local A/H1N1 outbreak [19], but Malaysians were particularly anxious about the pandemic [20]. The current study shows that emotional distress was relatively mild in China as few residents worried about being infected (25.1%). This phenomenon may also be related to the previous experience of the SARS epidemic, as well as the open epidemic information. A survey in Korean university showed that women perceived higher illness severity and personal susceptibility to A/H1N1 infection, which had been reconfirmed in our study [21]. Logistic regression analysis results suggested that women with higher educational level had higher perception of risk. As time went by, the knowledge about the main transmission route increased, but the risk perception of being infected in residents decreased, suggesting the positive effect of government policy regarding A/H1N1 infection prevention, as well as the promotion of the media.

The previous study presented various results of influencing factors on the the compliance with the preventive practices. The study in Saudi showed that older men with better education were more likely to take preventive practices [9]; female students in Korean washed hands more frequently during the peak pandemic period of A/H1N1 [21]; in another pandemic study in USA, younger people was found to have greater uptake of recommended behaviors but not for gender [16]. We found female with higher education took more precautionary behaviors, but office staffs and farmers took less comparing with students. While such differences could result from study population demographics, profound differences may also exist in the knowledge of A/H1N1 and the perceptions of recommended behaviors in those countries. Adjusting for the background factors, the multivariate logistic regression showed the possible relationship between knowledge and risk perception, knowledge and practices (odd ratios were 1.57 and 2.09, respectively), which indicated that good knowledge is important to enable individuals to have better attitudes and practices in influenza risk reduction. Similar findings were observed in other studies performed during A/H1N1 pandemic in Singapore [22] and during SARS pandemic in Hong Kong [13]. Therefore, it is important to focus on inculcating the correct knowledge to individuals as it will influence both attitudes and practices.

Injecting vaccination is an effective measure to prevent infectious disease [23]. In China, the seasonal influenza vaccination is not included in the national immunization program and must be purchased by recipients. Those who are above 60 years old, the pupil and children in kindergarten, and people with chronic diseases are recommended to get inoculation. Data provided by China CDC in 2009 showed that the immunization rate of the seasonal flu in Chinese population was below 2% [24], which was much lower than 7.5% in our study (P < 0.0001). This phenomenon is partly due to the state's prior vaccination strategy for population at high risk such as students, teachers, healthcare workers and people with chronic disease, as well as the confusion between seasonal flu vaccine and A/H1N1 vaccine in residents. People who couldn't access the A/H1N1 vaccine may take up seasonal flu vaccine as preventive behaviors. The A/H1N1 vaccine was not available in China until the middle of September 2009. All populations at high risk above three years old were invited for vaccination free of charge [25]. A survey among 868 European travelers showed 14.2% participants were vaccinated against pandemic influenza A/H1N1 [26], higher than 10.8% in our study (P < 0.01). Our study also showed students and health care workers were more likely to take up, which may be due to the prior vaccination strategy. Multivariate stepwise logistic regression analysis, which allowed us to adjust for background factors, further showed the perceived risk of infection and the knowledge about the main modes of transmission related to A/H1N1 vaccination were insignificantly, similar results seen in Lau's study[8]. Therefore, the vaccination rate of A/H1N1 is not expected to increase even if the virus becomes more prevalent or the knowledge of its transmission mode improved. Additionally, the behavior of taking up A/H1N1 vaccine was associated with perceptions of vaccine's safety and influence on daily life by A/H1N1 as well as the knowledge about the free vaccination policy and the state's initial vaccination strategy. This suggests that improving the safety of vaccine, the acceptability of side effect and the knowledge about the state's strategy related to A/H1N1 vaccination in residents may be helpful to promote A/H1N1 vaccination in the general population.

The cross-sectional telephone survey adopted in the study has some limitations. We were unable to interview the people who did not have phones and the depth of the questionnaire was largely limited because questions and pre-existing answers could not be too long and complex. In addition, the telephone response rate was 46.8%, which means more than half of the interviewees rejected or didn't finish the survey. It was impossible to compare the difference between respondents and non-respondents due to the lack of their basic information.

This A/H1N1 epidemic has not caused public panic yet, but the knowledge of A/H1N1 in residents is not optimistic as most of them confused the transmission route of A/H1N1. There are many factors influencing the KAP related to A/H1N1. Female with higher educational level had higher perceived risk of infection and took more precautionary behaviors. Public education campaign may take the side effects of vaccine and the knowledge about the state's vaccination strategy into account. The data collected in this survey could be used as baseline data to monitor public perceives and behaviors in the event of future outbreak of infectious disease in China.