Introduction

On December 7, 2022, China’s Comprehensive Group of the Joint Prevention and Control Mechanism released the “Ten New Measures” epidemic measure notice (CGSCC 2022), shifting the focus of COVID-19 prevention from infection control to health protection and severe disease prevention. The release of the “Ten New Measures” has made great changes in China’s epidemic prevention and control measures. The subsequent surge in infections led to increased demand for drugs, causing a supply–demand imbalance. Within a week of the publication of “Ten New Measures”, pharmacies and hospitals faced shortages due to panic buying by both infected and uninfected individuals. The study aims to examine the potential impact of infection information shared on social networks, information from physical social networks, and risk perception of COVID-19 on panic medication consumption following the release of the “Ten New Measures”. The research model is based on the "stimulus–organization–response" theory, aiming to explore the impact of these factors on individuals’ risk perception and consumer behavior during the epidemic. Despite being a very important factor in dealing with emergency management, the issue of panic consumption of drugs has not received enough attention from the academic community.

The questions addressed in this study are: Does knowing that social network members have been infected with the epidemic promote consumer panic consumption? Does browsing about infection rates on social media promote panic consumption? Does the perceived risk of infection with COVID-19 infection lead to panic consumption? Does the change in personal health level exacerbate panic consumption?

Literature review

Since the beginning of the COVID-19 pandemic in the last months of 2019, there has been a considerable amount of research investigating the influence of this public health crisis on consumer behaviors. When consumers received information about the severity of COVID-19, the level of their anxiety and panic drastically increased (van Esch et al. 2021). This pattern of behavior is already well known and explored in the area of social psychology as TMT theory (terror management theory) (Greenberg et al. 1986).

When consumers are in a state of increased anxiety or feel like they are not having control over the situation, they will seek behaviors that will give them the feeling of maintaining control (Gobrecht and Marchand 2023). One of those behaviors is panic buying and stocking items that may be perceived as useful during an anticipated crisis (Sim et al. 2020). Despite the fact that this pattern of response during the COVID-19 pandemic is noticed all over the world (Islam et al. 2021; Arafat et al. 2021; Li et al. 2021b; Billore and Anisimova 2021) researchers still do not have a uniform explanation about the trigger mechanisms behind this pattern of behavior.

In order to examine the relationship between the perceived risk of infection through social networks and panic consumption, our paper includes a presentation of relevant prior research on this topic.

Panic consumption

The consumption behavior of individuals, guided by the principle of maximizing utility, is commonly referred to as rational consumption behavior, contrasting with irrational consumption behavior. Compared with the objectivity, logicality, and efficiency emphasized by rational consumption, irrational consumption often has stronger subjectivity, experience, and non-binding characteristics (Billore and Anisimova 2021). It is obvious that a perceived lack of control and uncertainty throughout a specific emergency is the primary motivator of panic buying behavior, particularly in purchasing essential goods (Billore and Anisimova 2021).

Panic consumption refers to “the behavior of purchasing a large number of specific products or commodities due to the sudden worry about the upcoming shortage or price rise” (Islam et al. 2021). It is a common response of consumers to disasters, which usually occurs on the basis of consumers’ expectations of disasters or perceived disasters, such as what happened during the COVID-19 pandemic (Li et al. 2021b). The images of empty shelves circulated by the media provoke a desire to imitate such behavior, such as hoarding or stockpiling supplies (Prentice et al. 2022). With the emergence of COVID-19, information about the importance of health protection behaviors has been widely spread in the media around the world, triggering the demand for disinfectants, soap, masks, and gloves (Sherman et al. 2021). Panic consumption behavior is related to many undesirable consequences, ranging from disrupting economic stability to hindering the timely provision of supplies to those in dire need (Li et al. 2021b). Previous studies have focused on the relationship between COVID-19 and panic buying (Billore and Anisimova 2021; Islam et al. 2021; Sherman et al. 2021), the epidemic situation and channel switching behavior (Prasad and Srivastava 2021; Youn et al. 2021), pointing out that the psychological factors like fear and anxiety are among the underlying causes of panic consumption during an epidemic (Kemp et al. 2021; Loxton et al. 2020), cognitive factors (Herjanto et al. 2021; Septianto and Chiew 2021), demographic factors (Karpen and Conduit 2020; Peluso et al. 2021) and emotional factors (Eger et al. 2021). These irrational consumption behaviors not only increase the cost of social governance but also cause an additional burden on epidemic prevention and control, and bring greater anxiety and panic to the society (Loxton et al. 2020).

Social networks

Social networks as a term can be divided into physical or offline social networks and online social media networks. Physical social networks consist of close relationships with family members and friends, while online social networks consist of virtual and unsustainable relationships established by various online social networks (Yi et al. 2019). Physical social networks and online social networks represent different relationship characteristics (such as intimacy, contact frequency, and relationship duration) within an individual’s social network, so it is likely to affect personal health behavior through different mechanisms. Human beings are inherently social beings, and they possess a natural tendency to form connections and engage with others (Li and Meng 2023).

When there is a public crisis, the public’s discussion about the crisis events increases, thus affecting people’s thinking and rationality. Seeking relevant health information from the media but also from social circles is an important and common form of interpersonal interaction during the epidemic (Guo and Bai 2023). During the COVID-19 epidemic, consumer behavior became more and more complex and dynamic. It was the first global epidemic of such magnitude, and a significant majority of the population lacked experience and felt uncertain about how to respond and behave (Herjanto et al. 2022). During this period, social media played a significant role as an information channel. It was the most important, and often the only possible channel for social contact, but also the main source of information exchange as pandemic control measures advised the population to reduce person-to-person communication to only essential needs (Song et al. 2023). During the epidemic, the use of social media increased the impact of external stimuli on epidemic perception, often exacerbating panic levels among the population (Islam et al. 2021). In the Internet era, public crises usually spread to the public through social networks and the majority of people are receiving news through social media networks (Lee et al. 2023). Since the beginning of the epidemic crisis, it was almost impossible to access social media networks without seeing information related to COVID-19 and updates on infection rates (Li et al. 2021b). The unprecedented real-time information on COVID-19 at users’ fingertips can give them the tools they need to make smart decisions but also make them more anxious about what is to come, which may lead to panic buying or stockpiling of products (Naeem 2021). The infection of members in social networks, both physical and online, brings the epidemic situation closer to individuals, leading them to perceive an increased risk of contracting COVID-19. Previous researchers have found that the spread of COVID-19 has led to an increase in global media consumption (Billore and Anisimova 2021). Additionally, studies have shown that social media use significantly increases following preventive behaviors (Herjanto et al. 2021). Moreover, various studies (Oh et al. 2021; Zeballos Rivas et al. 2021; Liu et al. 2020) have reported a significantly higher level of perceived risk of infection in these groups. Notably, participants who had COVID-19-infected members in their social network, including relatives, friends, colleagues, and others in their social circles, exhibited an even higher level of perceived risk (Li et al. 2021b).

Literature limitations

In general, the existing literature has some limitations:

First, the conceptual model of the impact of social networks on health, and the existing literature pay less attention to the infection of social network members.

Second, when explaining the impact of social networks on health, there is a lack of consideration of infection risk perception. Without perceiving the risk of epidemic infection, the behavior of panic consumption cannot be triggered.

Third, there is a lack of sufficient research on the panic consumption of drugs that serve as crucial resources during an epidemic.

Research model and hypotheses development

Research model

The sor model illustrates the connection between external stimuli, which can be influenced by various factors, and the response of individuals, including their emotional state and subsequent behaviors (Lin et al. 2022). In sor theory, the stimulus is defined as the behavioral response of individuals affected and triggered by various environmental factors. In this theory, an organism is defined as any internal process that is affected by stimulation and drives the final reaction; and response is defined as the conversion of the initial two outcomes into behavior output (Alanadoly and Salem 2022).

Previous studies have explored the process of consumer behavior based on the stimulus–organism–response (sor) theory (Lin et al. 2022; Alanadoly and Salem 2022; Chopdar and Balakrishnan 2020; Guo et al. 2021). In the field of consumption, stimulus refers to various shopping atmospheres and website characteristics created by businesses. The organism is the emotional arousal of consumers due to the stimulation of businesses, while the response refers to consumer behavior under a series of internal and external stimuli (Lin et al. 2022).

Taking into consideration the situation in China after the release of the “Ten New Measures”, this paper constructs a research model based on sor theory (see Fig. 1). In our model, S (stimuli) is measured as infection information obtained from social networks, including infection information obtained from physical social networks and infection information obtained from online social networks.

Internal processes, emotional arousal, or organism reaction O are measured through risk perception, while R is measured by panic consumption, including conformity buying and uncontrolled self-medication.

Therefore, the model presented in this paper additionally explores the intermediary role of risk perception in the impact of infection information on panic consumption and analyses the regulatory role of health change in the impact of risk perception on panic consumption.

Fig. 1
figure 1

Theoretical framework based on sor theory

Full size image

Hypothesis development

From the perspective of information transmission function, social networks can play the role of accelerating information transmission before and after the occurrence of risks and disasters. Increasing the dissemination of information about the epidemic leads to individuals acquiring more knowledge and information about it. As a result, they are more likely to engage in effective self-protection behaviors.

The majority of the population lacks sufficient knowledge and experience on how to behave during the pandemic and public crisis situations, which makes them vulnerable to the influence of their social networks. After the release of the “Ten New Measures”, individuals in China were informed about the infection of their social network members through both physical social networks and online social networks. With the increasing infection rates, the risk perception of individuals should also become higher. When individuals have a high-risk perception, they may make purchase decisions because they observe the purchase behavior of others, or they may take medication on their own because of fear of infection. Therefore, considering the observed patterns in the relevant research, the following research hypothesis is proposed:

  • Hypothesis 1 (H1) Risk perception plays an intermediary role between infection information and panic consumption.

  • Hypothesis 1a (H1a) Risk perception plays an intermediary role between infection information and conformity buying.

  • Hypothesis 1b (H1b) Risk perception plays an intermediary role between infection information and uncontrolled self-medication.

The pandemic threatens everyone’s life, and individuals’ reactions to this threat depend on their perceptions of the likelihood that they might get infected and their tolerance of these uncertainties (Shahnawaz et al. 2022). With the emergence of COVID-19, information about the importance of good health behavior has been widely spread in the media around the world, triggering the demand for disinfectants, soap, masks, and gloves (Sherman et al. 2021).

When consumers perceive the seriousness of the threat and their vulnerability to the threat, they will more actively evaluate the response actions (Youn et al. 2021). Panic consumption is a kind of behavioral disorder, which is manifested in that consumers buy an abnormally large number of goods in order to avoid the possibility of future shortages (Herjanto et al. 2021). After the release of the “Ten New Measures”, on the one hand, the increased number of infections among social network members led to a higher risk perception of individuals, which will promote panic consumption; On the other hand, the decline of individual’s health status may cause individual’s concern about their own infection, especially for individuals whose physical condition has declined during the epidemic, making them more susceptible for conformity buying and uncontrolled self-medication. Therefore, the following research hypothesis is proposed:

  • Hypothesis 2 (H2) Health change can regulate the relationship between risk perception and panic consumption.

  • Hypothesis 2a (H2a) Health change has a regulatory effect on the relationship between risk perception and conformity buying, that is, the positive correlation between risk perception and conformity buying is stronger when the health level decreases than when the health level increases.

  • Hypothesis 2b (H2b) Health change have a regulatory effect on the relationship between risk perception and uncontrolled self-medication, that is, the positive correlation between risk perception and uncontrolled self-medication is stronger when the health level is decreased than when the health level is improved.

Methodology

The questionnaire that was used for the purpose of this research consists of a demographic information part and four scales that were used to measure variables of interest in this research. Three scales were derived from the previous relevant studies on this topic and they were slightly expanded or adjusted to the aims of our study. Each scale in this study measures how often participants engage in the behaviors described in the statements following the release of the “Ten New Measures”. The Likert five-level scale was used to collect answers, with the value from 1 to 5 (from lower to higher) representing: “very inconsistent”, “inconsistent”, “average”, “consistent” or “very consistent”. More precisely, the variables of interest in this study are measured as follows:

  1. (1)

    Panic consumption is measured by conformity buying (cb) and uncontrolled self-medication (sm). Conformity buying is measured with six items that measure the influence of different instances in the surroundings such as experts, relatives, and friends, developed by Li and his colleagues for the purpose of their research (Li et al. 2021a). Uncontrolled self-medication was measured through self-developed items related to preventive medication, repeated medication, and increased dosage of regularly taken medication. Cronbach’s coefficient for the uncontrolled self-medication in the questionnaire was \(\alpha \ 0.84\), and for conformity buying \(\alpha \ 0.88.\).

  2. (2)

    Infection information is a term that pertains to previous research concerning information about infection rates (Bian et al. 2021). In our questionnaire, we examine whether participants’ dominant information source regarding infection rates originates from online or offline social networks. We assessed this using seven items. Four questions of the scale measure the infection information obtained from physical social networks (pi) such as through contacts with co-residents, relatives, and friends about infection rates in their surroundings. Three questions measure the infection information obtained from online social networks (oi), such as through social media or online communication channels like WeChat, Weibo, or short video platforms.

  3. (3)

    Risk perception (rp) is measured through the scale previously developed by Dryhurst and his team (Dryhurst et al. 2022). The scale consists of five items and measures individuals’ perception of the risk of COVID-19 infection for themselves, their family members, or others in their surroundings over the next 3 months. Cronbach’s \(\alpha \) coefficient for the risk perception scale in the questionnaire amounted to \(\alpha \ 0.80.\)

  4. (4)

    Health change (hc) is a crucial variable in this survey as it pertains to the potential changes in participants’ health during the pandemic, which may influence their perspectives and perception of the situation. To assess whether participants’ health underwent changes throughout the epidemic, particularly following the release of the “Ten New Measures”, we incorporated questionnaire items to measure this aspect of their personal experience with COVID-19 infection.

Sample and data collection

For the data collection procedure, we used an online survey platform (wyx.cn). Before entering the study, informed consent was obtained from each participant after they were initially informed of the survey’s purpose. Following the survey, participants were provided with contact information so they could learn more about the study’s objectives. Since all participants were financially reimbursed for their participation, the research procedure could not be conducted anonymously. However, all personal data were deleted from our records upon completion of the experimental procedure.

The questionnaire was released in December of 2022 when COVID-19 infections were reaching their peak. In the first phase, 593 answers were collected, of which 408 were included in the final analysis with an effective rate of 68.8%. A certain number of answers had to be excluded from the final data set as they contained invalid or missing values. Participants are coming from the general population, so the demographic characteristics of the participants are various as can be seen in Table 1. When it comes to the gender distribution of the sample 66.2% of participants are male, and the proportion of those aged 18 to 40 years old is more than 90%, so it is important to notice a lack of participants older than 40 years old. The highest achieved education levels of participants are mainly junior college and undergraduate degrees, accounting for 22.3% and 58.1% respectively. When it comes to the monthly income numbers, results on our sample have shown a normal or Gaussian distribution as 37.3% of participants’ incomes are in the range between 5000 and 8000 Renminbi, which reflects the average monthly income. Other monthly income ranges are equally distributed. In the period between the release of the “Ten New Measures” and the completion of the questionnaire, 28.7% of the respondents had already received a diagnosis or were found to be asymptomatic cases of COVID-19.

Table 1 Socio-demographic characteristics
Full size table

Reliability, validity, and correlation analysis

This study conducted reliability and validity analysis using SPSS version 24.0. To determine if the data are appropriate for factor analysis, which helps identify underlying patterns and relationships among variables we used Kaiser–Meyer–Olkin coefficients (kmo) in our calculations. The sample is sufficient if the kmo values are between 0.8 and 1.0. kmo, values between 0.6 and 0.69 are mediocre, whereas those between 0.7 and 0.79 appear to be average (Shrestha 2021). The Kaiser–Meyer–Olkin (kmo) value for all variables in our survey was higher than 0.7, which is considered an acceptable score confirming the correlation between the variables. Therefore, factor analysis was conducted, and Cronbach’s \(\alpha \) coefficient was calculated to assess the internal consistency reliability of each variable. Generally speaking, a sufficient Cronbach’s \(\alpha \) value is one that is greater than 0.7. Such values of \(\alpha \) indicate that the assessed items are significantly related (Lavrakas 2008). The Cronbach’s \(\alpha \) coefficients of all dimensions were \(>0.7\), indicating internal consistency and high reliability.

The square root of the average variance extracted (ave) value of each variable is mostly greater than the correlation coefficient between the variables, indicating that the difference validity between the variables is high. The correlation analysis has laid a good foundation for future research on the variables of our interests.

Table 2 Regression analysis of risk perceptions
Full size table

Results

Risk perception

Table 2 presents the results of the stepwise regression analysis on variable risk perception. The predictor variables included in the regression model are gender, age, education, monthly income, infection information obtained from online social networks (oi) and infection information obtained from physical social networks (\(\pi \)). The results have shown that variables infection information coming from both physical (M2) and online social networks (M3) significantly contribute to individuals’ risk perceptions when they are separately included in the model. However, when both predictors are included together in the regression model (M4), infection information obtained from physical social networks \((\beta = 0.29,\ p < .01)\) is proven to be a better predictor of risk perception scores than online social networks \((\beta = 0.14,\ p < .01)\). In the analysis, demographic characteristics such as gender, age, education level or monthly income, were not found to be significant predictors \((all \ p > .05)\) of risk perception. The overall model fit was statistically significant \((F= 8.796,\ p < 0.01)\), indicating that the considered predictors together accounted for a significant amount of variance in risk perception. The proposed regression model explained \(11.6\%\) of the variance in conformity buying \((R^2 = 0.116,\ p < 0.01)\).

Panic consumption

As previously mentioned, panic consumption in this paper is measured through variables conformity buying and uncontrolled self-medication. Results in Table 3 pertain to the variable conformity buying as the dependent variable, whereas the proposed models in Table 4 are associated with uncontrolled self-medication as the dependent variable.

Conformity buying In the stepwise regression analysis predicting conformity buying, several predictors were considered. The predictor variables included in the model are gender, age, education, monthly income, infection information obtained from physical social networks (oi), infection information obtained from physical social networks (\(\pi \)), health change (hc), risk perception (rp), and the interaction effect between health change and risk perception (hc \(\times \) rp).

Among the individual predictors, risk perception (rp) has proven to be a significant and best predictor of conformity buying among all variables in model \((\beta = 0.79,\ p < 0.01)\), indicating that higher levels of risk perception positively impact the increase in conformity buying behaviors.

The variable health change (hc) did not show a significant relation with conformity buying \((\beta = -0.02,\ p > 0.1)\), but a significant negative correlation between conformity buying and the interaction effect of health change and risk perception (hc \(\times \) rp) was found \((\beta = -0.26,\ p < 0.05)\). This suggests that scores on the variable risk perception affected the connection between health change and conformity buying, and these two predictors together impact the results on the variable conformity buying. Regarding the variables of infection information, physical social networks (\(\pi \)) had a significant positive association with conformity buying \((\beta = 0.27,\ p < 0.01)\), suggesting that the information about infection rates coming from physical social networks is associated with higher levels of conformity buying. On the other side, infection information coming from online social networks (oi) did not show a significant association with conformity buying \((\beta = 0.027,\ p > 0.1)\). Surprisingly, demographic variables such as age, gender, education, and monthly income also are not significant predictors of the score on variable conformity buying \((all\ p > 0.1)\). The overall model fit for the regression analysis on the variable conformity buying was statistically significant, as indicated by the F-values \((F = 24.14,\ p < 0.01)\), suggesting that the included variables collectively contributed to explaining the variance of the dependent variable. The proposed regression model explained \(35.3\%\) of the variance in conformity buying \((R^2 = 0.35,\ p < 0.01)\) with all considered variables included in the model.

Table 3 Regression analysis of conformity buying
Full size table

Uncontrolled self-medication The regression analysis of scores on uncontrolled self-medication includes the same set of variables as in conformity buying. The results have shown that gender was a significant predictor of the uncontrolled self-medication \((\beta = -0.153,\ p < 0.01)\), indicating that women are more prone to take drugs on their own than men. Other demographic characteristics; age, education level and monthly income, had no significant predictor value when it comes to uncontrolled self-medication \((all\ p > 0.1)\). Regarding the transmission of infection information, in the initial model (M1) where only two variables related to infection information and demographics were considered, infection information obtained from physical social networks (pi) emerged as a significant predictor of uncontrolled self-medication \((\beta = 0.14,\ p < 0.01)\). In the same model, infection information obtained from online social networks (oi) was not a significant predictor \((\beta = -0.06,\ p > 0.1)\). But, in the final model (M5), when all variables are included in the regression analysis, infection information obtained from physical social networks (\(\pi \)) was not a significant predictor of uncontrolled self-medication anymore \((\beta = -0.01,\ p > 0.1)\), while infection information obtained from online social networks (oi) had a negative significant correlation with uncontrolled self-medication \((\beta = -0.12,\ p < 0.01)\), indicating that receiving information about infection from online social networks cause a decrease in the amount of uncontrolled self-medication. The results of conducted regression analysis have shown that the significant predictor of uncontrolled self-medication is also the variable (rp) risk perception \((\beta = 0.33,\ p < 0.05)\), implying that the higher risk perception of infection with COVID-19led to a higher uncontrolled self-medication. As expected, health change (hc) also had a significant negative correlation with uncontrolled self-medication \((\beta = -0.09,\ p < 0.05)\), indicating that the worsened health state led to an increase in uncontrolled self-medication. The interaction between risk perception and health change (hc \(\times \) rp) on uncontrolled self-medication was not found to be significant \((\beta = 0.11, p > 0.1)\). The overall model fit for the regression analysis variable uncontrolled self-medication was statistically significant, as indicated by the F-values \((F = 13.61,\ p < 0.01)\), suggesting that the considered predictors collectively contributed to explaining the variance of the dependent variable. The proposed regression model explained \(23.5\%\) of the variance in uncontrolled self-medication \((R^2 = 0.24,\ p < 0.01)\) when all considered variables were included in the model.

Table 4 Regression analysis of uncontrolled self-medication
Full size table

Moderated mediation effect analysis

Based on the results of the regression analysis presented earlier, it was found that when infection information is obtained through physical social networks, it has a positive effect on panic consumption by influencing risk perception. To further examine the mediating role of risk perception regulated by health change in the relationship between infection information and panic consumption, an intermediary model test procedure was conducted. The moderated mediation model test is conducted using the bootstrap method and 5000 sample groups at a 95% confidence interval. The results indicate that the variable risk perception has an intermediary effect in the impact of infection information on panic consumption. Considering the mean value and the mean value plus or minus one standard deviation, the low, medium, and high levels of health change are stratified, and the intermediary effect of risk perception under different health change levels is analyzed. The analysis reveals a significant indirect effect of infection information on conformity buying through risk perception, with the indirect effect gradually decreasing as health levels improve. However, the moderated mediation effect of infection information on uncontrolled self-medication is not significant. The analysis reveals that infection information significantly affects both conformity buying and uncontrolled self-medication intake through risk perception, with a notably greater indirect effect on conformity buying. Changes in health play a regulatory role in the impact of risk perception on conformity buying but do not significantly regulate its impact on uncontrolled self-medication. The decline in health levels leads to increased concern, promoting conformity buying driven by risk perception. Consequently, the decline in health levels during an epidemic does not significantly enhance the impact of risk perception on uncontrolled self-medication. However, uncontrolled self-medication intake may be influenced by other factors beyond health change.

Discussion and conclusion

This study aims to propose a research model based on sor theory, collect data through questionnaires, and investigate the factors influencing panic consumption of drugs. Combined with the situation of epidemic prevention and control measures in China, this paper measured panic consumption through conformity buying and uncontrolled self-medication and analyzed the impact of infection information obtained from physical social networks and online social networks on panic consumption. The role of risk perception in mediating the impact of infection information on panic consumption is revealed, indicating the differential influence of risk perception in this context.

Research findings

The main research conclusions are as follows: First, infection information obtained from both physical social networks and online social networks has a significant positive impact on risk perception while infection information obtained from physical social networks has a higher individual impact. Further, infection information obtained from physical social networks and online social networks has asymmetric effects on conformity buying and uncontrolled self-medication. infection information obtained from physical social networks has a positive significant impact on conformity buying, and not significant on uncontrolled self-medication when all the variables are included in the regression model. On the other side, infection information obtained from online social networks has a significant negative impact on the uncontrolled self-medication, while its impact on conformity buying is not significant. Furthermore, in the intermediary effect test, relevant variables have asymmetric effects on conformity buying and uncontrolled self-medication. Infection information obtained from physical social networks affects conformity buying and uncontrolled self-medication through risk perception. Compared with the uncontrolled self-medication, infection information obtained from physical social networks and risk perception can cause a higher level of conformity buying. Lastly, in the impact of risk perception on conformity buying, health change played a significant negative regulatory effect, but in the impact of risk perception on the uncontrolled self-medication, health change did not play a significant regulatory effect.

Contributions to the literature

First, although some scholars (Billore and Anisimova 2021; Chen et al. 2022) have conducted research on risk perception and panic consumption in the context of the COVID-19 epidemic, there are very few studies on the panic consumption of drugs in the context of the epidemic. The research context of this article is the issue of panic consumption of drugs after the release of “Ten New Measures” in China. The panic consumption at this stage is not only reflected in the purchasing process but is also reflected in the higher consumption of medicine or in increased dosage of medication. The last mentioned may exacerbate the contradiction between the supply and demand of medicine, while in the long term, it may also have a potential negative impact on physical health.

Therefore, this study measures panic consumption by conformity buying and uncontrolled self-medication, which is consistent with the problem of panic consumption of medicine caused by the adjustment of COVID-19 policy (“Ten New Measures”). Also, scholars have found that social networks and social influences have an impact on panic consumption during the epidemic period (Naeem 2021; Wang et al. 2023; Yuen et al. 2021) but there are few studies on the impact of social networks on medicine consumption in the context of the epidemic. This study does not focus on information about infection rates coming from a social network, but rather on the information about infection among social network members, revealing the differential impact of infection information on panic consumption in two types of social networks, physical and online. Therefore, it is different from the research findings on the impact of online social networks on panic consumption (Naeem 2021). This research study postulates that panic consumption is significantly and more strongly influenced by infection information disseminated through physical social networks when compared to infection information obtained from online social networks. Also, in the previous work many scholars have researched the impact of perceived severity (Yuen et al. 2021) and perceived scarcity (Naeem 2021) on panic consumption in the context of the epidemic, but these studies were focused on products of daily use and not on drugs like it is the case in our study.

Implication for practice

This study found that risk perception significantly affects panic consumption and plays an intermediary role in the impact of infection information on panic consumption. Therefore, the risk perception could be reduced from the two aspects. On the one hand, the government should open more fever clinics, expand the resources of critical care, provide online consultation and medication guidance through Internet diagnosis and treatment services, and ensure the medical needs of the people during the epidemic period. On the other hand, the government should use social media to release up-to-date scientific information on the epidemic infection through authoritative experts in a timely manner, promoting a better understanding of epidemic infection and helping individuals to master scientific response strategies. In addition, false information on social networks that amplify the risk of epidemic infection should be promptly refuted and controlled, so as to reduce the perceived risk of individuals.

Additionally, this study found that the decline in health levels aggravates the impact of risk perception on panic consumption. When individuals get infected, it intensifies their concern about their own health prompting them to stockpile medications. Thus, in the state of higher infection rates, higher panic consumption rates are to be expected. Lastly, as presented in our results, after the release of “Ten New Measures”, risk perception among the participants increased leading to a higher level of panic consumption. In light of this understanding, authorities and pharmaceutical distributors should evaluate the existing stock of medical products prior to issuing any statements on the epidemic situation, such as the “Ten New Measures” announced in China in December 2022. This approach is essential to mitigate the discrepancies between supply and demand. Likewise, authorities should also promote drug use guidelines promptly in the non-epidemic state so that the population can be better prepared for the potential situation of a future large-scale epidemic and to hinder large-scale herd purchase of medicine during the pandemic.

Limitations and further research

An interesting observation from our findings was that demographic factors had either no significant impact or only a negligible effect (such as gender on uncontrolled self-medication) on the variables of interest in our models. There is a potential for studies to further investigate the relationship between demographic factors and variables of panic consumption and risk perception. Additionally, despite having samples of participants with various backgrounds, there is to notice a lack of elderly participants which is somewhat of a limitation as they are a high-risk population considering the negative health effects of the COVID-19 pandemic. Lastly, the regression models we presented could explain a significant number of variances of dependent variables, but \(R^2\) values of the proposed models were still relatively low, suggesting that future studies could benefit from the inclusion and consideration of additional variables.