Introduction

Food safety is a global health goal. Foodborne diseases represent a growing public health problem in developed and develo** countries [1]. Global estimates that 31 foodborne hazards cause 600 million foodborne illnesses and 420,000 deaths annually, resulting in the loss of 33 million healthy life years [2]. According to Foodborne Diseases Surveillance Network, a total of 2,401 foodborne diseases occurred and resulted in 21,374 cases and 139 deaths in the 29 provinces of mainland China in 2015 [3]. Moreover, cases of foodborne diseases were often under-reported, especially in develo** countries [4].

Food safety has been of concern in university settings in recent years. University students are one of the high-risk population groups for food poisoning, who have inadequate knowledge [5,6,7] and risky food safety practices. University students typically eat out [8], consume takeaway food [9] and have unhealthy food handling [10]. Normal and medical university students belong to a population with unique features. Normal universities provide teacher education in China, in which various types of teachers were trained. Food safety cognition and practices of teachers and doctors are beneficial to their food safety incidence prevention and are expected to play important roles in health education and promotion after their graduation [6].

The World Health Organization stated that food safety education is vital in eliminating or reducing food contaminants and preventing micro-organism growth at levels that cause disease [11]. Some food safety intervention programs were conducted targeting food service employees [12,13,14] and students [15, 16], and traditional education methods were often used [17, 18]. These methods included providing reading materials (e.g. booklets and leaflets), conducting lectures and presentations and distributing posters [19]. A previous study demonstrated that methodology and approach adopted are important for a successful food safety training programme [20]. The limited effectiveness of traditional health education [21, 22] leads health education and promotion researchers worldwide to explore effective and innovative ways, which attempt to increase the efficacy of their interventions based on the worldwide web and other digital media [23].

One of the leading social networks worldwide, WeChat, developed by the Chinese company Tencent, placed fifth in the number of active users and had over 1.1 billion monthly active users in the first quarter of 2019 [24]. In accordance with the statistics provided by the China Internet Network Information Center in 2019 [25], the percentage of WeChat users reached 83.4% in China. Most of its users were between the ages of 20 and 29 by the end of December 2018. Like Facebook, Twitter and Instagram, WeChat offers a free instant messaging application for smartphones that enables the exchange of voice, text, pictures, videos and location information via mobile phone indexes [26]. WeChat official account is based on a new functional module. WeChat users can register an official account, which enables them to push feeds, interact with one another and provide subscribers with service. In addition, subscribers can read messages and communicate with others through these official accounts [27]. At present, WeChat, as a cost-effective and peer-to-peer supported educational tool, has been used for conducting health education or promotion to modify behaviours [27,28,29,30]. However, concerns exist about reliability and quality control of disseminated information via social media, as well as concerns about the intervention effects on promoting healthy behaviours [31,32,33]. Understanding the effect of WeChat on users is important as it gains popularity as a health intervention platform.

At present, most intervention strategies for improving food safety cognition and practices are mainly based on traditional education methods. Previous studies on social media and food safety were targeted at communication of food safety risks or public opinion on the Internet regarding food safety [27, 34, 46, 47], and medical students may have no time and energy to participate in this intervention. Thus, new methods to mobilize the enthusiasm and increase the participation of university students should be explored and attempted in a future intervention study, such as sending links or emails to invite participants to view relevant content [48]. In addition, the duration of social media intervention should avoid the period when university students would be busy with many examinations.

Secondly, part of intervention information through articles published in "Yingyangren" WeChat official account may not be appropriate for university students. "Yingyangren" WeChat official account is a relatively experienced platform for delivering health knowledge, university students in this intervention are a part of this official account followers. Some contents of the articles provided are related to popular issues to attract interest of official account followers, but the relevance to university students is under considered, such as the veterinary drug residue of meat and pesticide residues in vegetables; these topic can be hardly put into practice by the Chinese university students who mostly live a campus life. Information intervention may be invalid. WeChat official account should be set up specifically in future intervention for target population audience. In addition, KAP model was used to evaluate the intervention effects in this study. In order to ensure the reliability and validity of the KAP questionnaire, the items of questionnaire were selected. Hence, the final questions used to measure the intervention effects were not exactly kept in line with the intervention materials/science articles; this could be the one of the reasons for the limited intervention effects.

Thirdly, diversified health information acquisition amongst university students [49] may be another reason for the limited effects. In this study, popular science articles released by the "Yingyangren" WeChat official account would be re-tweeted via WeChat moment, QQ or Micro-blog to increase accessibility amongst the intervention group. However, during the process, the participants in the control group may also obtain the intervention information indirectly. Moreover, results of this study showed that more than half of the participants in the control group obtained food safety-related knowledge via other social media or network platforms. In addition to the provided intervention platform and information, the participants can also obtain food safety-related information through other channels, which may cause a considerable improvement in food safety knowledge and practices in the intervention and control groups after the intervention and the no difference between-group findings in this study.

Intervention strategies of social media could enhance the success rate, such as the integrated use of discussion boards, learning modules, tailored feedback and interactivity [50, 51]. However, in this study, the WeChat official account was used to release food safety-related popular science articles, and learning modules were mainly intervention strategies for participants, whilst other functions of WeChat are not utilized efficiently in this study. This factor may be considered in analysing the limited intervention effects. In future studies, discussion boards, tailored feedback and interactivity of the WeChat official account should be utilized efficiently to enhance the success rate. Moreover, using social media as part of a complex intervention, which can combine the WeChat official account for online food safety education and offline lectures or food safety-related compulsory courses, could be conducted amongst university students.

The results of this study showed that intervention materials had a certain degree of readability and effectiveness. In addition, the participants had a relatively high level of satisfaction with the "Yingyangren" WeChat official account for conducting food safety education intervention. Most participants trusted the food safety-related popular science articles released by "Yingyangren" WeChat official account and agreed that the information could help to improve their food safety knowledge and correct their inappropriate behaviours. However, subjective assessment was not in accordance with the intervention results. The possible explanation for the inconsistent results could be that university students is relatively optimistic and may exaggerate the effects intervention, and the questions used to measure the intervention effects is somewhat difficult for them. Additional studies on how to efficiently use the WeChat official account to improve the food safety knowledge and correct inappropriate food safety behaviours of university students should be conducted.

This study has certain limitations. Firstly, this study did not design a targeted educational program for the student audience, and "Yingyangren" WeChat official account was not specifically established for this intervention group; part of intervention information may not be appropriate for university students; the KAP questionnaire was not exactly kept in line with the intervention materials. Future intervention by social media needs to be strengthened in these three aspects. Secondly, the intervention duration might not be enough. Practice change needs regular long-term education. One systematic review showed that the duration of social media intervention ranged from three months to two years [32]. However, the duration of this study was two months. The study duration could be increased to examine the intervention effects in future studies. Moreover, evaluating the food safety KAP during the two-month intervention process should be considered instead of just evaluating the KAP before and after the intervention, such as conducting an assessment after completing the three or four-time interventions. Thirdly, interaction characteristics in social media are one of the most common features [32, 52, 53], such as message boards and consulting section in this study. However, very few participants expressed their own opinions or raised questions about released food safety-related popular science articles. How to efficiently use the essential interaction characteristics should also be studied. Fourthly, this study relied on self-report, which may introduce bias caused by dishonesty, measurement flaws or social desirability bias. Fifthly, our intervention was implemented in small sample. Generalizability to larger units would be necessary. Finally, usage of Internet interventions was typically low, and high attrition rates are one of the possible reasons [32]. Similarly, high attrition rate in this study could introduce bias into the results, although no difference exists in the socio-demographic characteristics between the intervention and control groups before and after the intervention. Challenges of adherence and kee** the participants engaged, incentive motivation and end-user engagement during the development of the intervention could be attempted in future research to decrease the attrition [54, 55].

Conclusion

The WeChat official account intervention had a limited effect on improving the food safety KAP amongst university students. This study was an exploration of food safety intervention using the WeChat official account; valuable experience can be provided for social media intervention in future study. Given that university students are the key population for food safety intervention and social media has become the main method for them to obtain information, powerful trials and meta-analyses are required to explore how to efficiently use the WeChat official account intervention on food safety health education and how to improve the intervention effects in future studies.