Factors Associated with Refraining from Purchasing Foods Produced in Affected Areas after the Fukushima Daiichi Nuclear Power Station Accident

After the accident at the Fukushima Daiichi Nuclear Power Station, food products from the areas affected by the accident suffered reputational damage worldwide. Therefore, the present study aimed to identify the factors associated with people refraining from purchasing foods produced in affected areas to avoid radioactive materials. The study also aimed to clarify the time trends for the avoidance of foods produced in Fukushima Prefecture. We used data from “A survey on consumer awareness of reputational damages” conducted by the Japanese Consumers Customer Agency and implemented statistical analysis. Even if the year since the accident differed, “living with children”, “knowing detailed information about food inspections”, and “not being able to accept radiation risk even if the level is below the standard” were commonly associated. Not only did this study reveal that some people’s risk perceptions are fixed even when new knowledge is provided, but it also suggests that the implementation of food inspection can promote reputational damage. Additionally, the avoidance of Fukushima food products was found to decrease as time passed after the Fukushima accident. The results of this study may help develop countermeasures against reputational damage to food products after future nuclear disasters.


Introduction
On 11 March 2011, an accident occurred at the Fukushima Daiichi Nuclear Power Station, which caused a mass of radionuclides to be released into the atmosphere, contaminating the surrounding environment, including the land and ocean [1]. After the accident, the Japanese government set a provisional value for food and restricted the distribution of food products that exceeded 500 Bq/kg to reduce and prevent internal radiation exposure among the public [2]. As a result of this restriction, the internal radiation exposure of many Japanese people was suppressed [3]. However, Fukushima Prefecture products suffered reputational damage, and there has been a growing domestic push to avoid purchasing products from the prefecture [4]. Afterward, the Japanese government set a standard value for food at 100 Bq/kg to ensure safety [2], but this did not ease the trend of people refraining from purchasing food from the Fukushima Prefecture due to rumors. At present, although food products from the Fukushima Prefecture and other affected areas that exceed the standard value are extremely limited [5], nine countries, (China, Korea, Indonesia, United Kingdom, Iceland, Norway, Switzerland, Liechtenstein, and Russia) and five regions (Macau, Hong Kong, Taiwan, Polynesia, and the European Union) restrict the import of these products [6]. This cannot be based on science as the food is distributed and exported based on strict standard values [7]. The reputational damage of Fukushima products has been spreading, not only in Japan but also in the world, and it can be said that this damage is ongoing. The term "reputational damage" is defined in Japan as "economic damage caused by the fact that people consider foods, products, and land that are supposed to be safe to be dangerous and stop consuming or visiting them due to the extensive media coverage of an incident, accident, environmental contamination, or disaster" [8]. This kind of reputational damage has also occurred in Japan during the bovine spongiform encephalopathy problem, which led to the failure of US beef sales [9].
Since 2013, after the accident occurred, the Consumer Affairs Agency of Japan has been conducting "A survey on consumer awareness of reputational damage" [10] and has been investigating the trends of awareness of Fukushima products in Japan. However, this was a single-year assessment that did not look at trends over time. It has also not identified factors associated with people refraining from purchasing foods produced in affected areas to avoid radioactive materials. Therefore, we considered that, by clarifying the factors associated with the awareness and avoidance of foods containing radioactive materials immediately after the Fukushima accident to the present, it would be possible to examine future measures against reputational damage to food. This could also help develop countermeasures against reputational damage to food products after future nuclear disasters. In the present study, as well as factors associated with risk perception generally, we hypothesized that sex and family members living in the same household would be associated with refraining from purchasing foods produced in affected areas to avoid radioactive materials.
In the present study, we aimed to identify factors associated with this behavior. We also aimed to clarify the trends of these perceptions over a long period of time.

Study Design
This study employed a repeated cross-sectional study design.

Participants
A total of 5126 men and women in their 20s to 60s who live in the areas most affected by the accident at the Fukushima Daiichi Nuclear Power Station (Iwate, Miyagi, Fukushima, and Ibaraki Prefectures) were included in this study. Included also was the same demographic living in the main destination prefectures for agricultural, forestry, and fishery products from the affected prefectures (Saitama, Chiba, Tokyo, Kanagawa, Aichi, Osaka, and Hyogo Prefectures).

Data Source
We submitted a request to the Consumer Affairs Agency of Japan to provide the raw data for the survey. The request was accepted. A summary of the survey results can be obtained from the Consumer Affairs Agency's web page [10].

Question Items
The question items were developed by the Consumer Affairs Agency of Japan, aiming to "Continuously survey the reasons why consumers are refraining from purchasing agricultural, forestry, and marine products from the disaster-affected areas, and to utilize this information in future efforts to counter reputational damage and promote consumer understanding, including the content of explanations in risk communication", [10]. Some of the question items were reduced the longer the amount of time from the accident grew.
The question items included demographic factors such as "Sex", "Age", "Living area", "Marital status", "Living with infants", "Living with junior high school students", "Living with high school students", and "Living with elderly persons over 60 years old".
The items also include knowledge about food inspection for radioactive materials, such as: "In municipalities where foods exceeding the standard values have been confirmed, the shipment, distribution, and consumption of such foods are stopped"; "Inspection of radioactive materials in food is implemented in 17 prefectures around the eastern region"; "Based on the guideline by the nuclear emergency response headquarter, inspection plans are formulated in municipalities, after which inspection is conducted"; "The results of an inspection are released on the web page of the Ministry of Health, Labor, and Welfare (MHLW)"; "In the case that the inspection result exceeds the screening level, inspection by Germanium (Ge) detector is implemented"; "In inspection plans, the contamination of agricultural land and the inspection results for food is stated"; "I do not know that inspection of food is conducted". Furthermore, "Radiation risk perception", and "Avoidance of purchasing food produced in Fukushima Prefecture" were included in the questionnaire.

Statistical Methods
Those who answered "Yes" to the question item "Do you refrain from purchasing foods produced in the affected areas to avoid radioactive materials?" were set as criterion variables. Additionally, we selected and analyzed the common question items from the first to 14th survey.
We used a chi-squared test to compare the independent categorical variables and selected the p < 0.1 variables as explanatory variables. After the chi-squared test, we used binomial logistic regression analysis to clarify the variables' association with the avoidance of purchasing foods produced in the Fukushima Prefecture after the accident. A maximum likelihood method was used for variable selection. All data were statistically analyzed using SPSS Statistics (version 27.0; IBM, Tokyo, Japan). Statistical significance was set at p < 0.05.
Regarding the avoidance of foods produced in Fukushima Prefecture, the changes over time from the first survey to the last survey (the 14th survey) were summarized. Additionally, factors associated with people refraining from purchasing foods produced in the affected areas to avoid radioactive materials after the accident were compared between the results of the first survey and last survey.

Chi-Squared Test for the First Study
When compared to "Refraining from purchasing foods produced in affected areas to avoid radioactive materials" (n = 1443), "Not refraining from purchasing foods produced in affected areas to avoid radioactive materials" (n = 2088) showed differences in "sex" (p < 0.001), "age" (p < 0.001), "residential area" (p < 0.001), "marital status" (p < 0.001), "living with infants" (p < 0.001), "living with elementary school children" (p = 0.016), "knowing that inspection of radioactive materials in food is implemented in 17 prefectures around the eastern region" (p = 0.001), "knowing that in case the inspection result exceeds the screening level, inspection by Ge detector is implemented" (p = 0.019), "not knowing that food inspections are conducted" (p = 0.048), and "risk perception" (p < 0.001) ( Table 1). , and an affirmative response to "Inspection of radioactive materials in food is implemented in 17 prefectures around eastern region" (OR, 1.432; 95% CI, 1.174-1.748; p < 0.001) were risk factors for refraining from purchasing foods produced in affected areas to avoid radioactive materials. Regarding risk perception, an affirmative response to "Cannot accept even less than the standard value" (OR, 4.103; 95% CI, 3.265-5.156; p < 0.001) was a risk factor. On the other hand, an affirmative response to "Not caring about risk" (OR, 0.31; 95% CI, 0.233-0.413; p < 0.001) was a protective factor against refraining from purchasing foods produced in affected areas to avoid radioactive materials ( Table 2).

Chi-Square Test for the 14th Study
"Not refraining from purchasing foods produced in affected areas to avoid radioactive materials" (n = 2536) showed differences in "age" (p = 0.002), "residential area" (p = 0.006), "marital status" (p = 0.01), "having cohabitants" (p = 0.031), "living with elementary school children" (p < 0.001), "knowing that in municipalities where foods exceeding the standard values have been confirmed, shipment, distribution and consumption of the foods are stopped" (p < 0.001), "knowing that, based on the guideline from the nuclear emergency response headquarters, inspection plans are formulated in municipalities, after which inspection is conducted" (p = 0.001), "knowing that the results of an inspection are released on the web page of MHLW" (p < 0.001), "knowing that in case the inspection results exceed the screening level, inspection by Ge detector is implemented" (p < 0.001), "knowing that in inspection plans, contamination of agricultural land and the inspection results of food are stated" (p < 0.001), "not knowing that food inspections are conducted" (p = 0.048), and "risk perception" (p < 0.001) when compared to "refraining from purchasing foods produced in affected areas to avoid radioactive materials" (n = 728) ( Table 3).  ; p = 0.022) were risk factors for refraining from purchasing foods produced in affected areas to avoid radioactive materials. Furthermore, "being a widow" (OR, 4.455; 95% CI, 1.673-11.858; p = 0.003), "living with elementary school students" (OR, 1.738; 95% CI, 1.316-2.897; p < 0.001), and "cannot accept even less than the standard value" (OR, 3.307; 95% CI, 2.50-4.373; p < 0.001) were also risk factors for refraining from purchasing foods produced in affected areas to avoid radioactive materials. On the other hand, "not knowing that inspection of food is conducted" (OR, 0.65; 95% CI, 0.525-0.805; p = 0.006) was a preventive factor against refraining from purchasing foods produced in affected areas to avoid radioactive materials (Table 4).  (Figure 1).

Discussion
In the present study, we investigated factors associated with the public's perception regarding refraining from purchasing foods produced in affected areas to avoid radioactive materials and avoidance of foods produced in the Fukushima Prefecture and clarified the trend over time through 14 surveys.
In the first study, "Refraining from purchasing foods produced in affected areas to avoid radioactive materials" was associated with the following factors: residential area, age, sex, living with infants, knowledge about actual food inspection, and risk perception. On the other hand, in the 14th study, it was associated with the following factors: residential area, marital status, living with elementary school students, knowledge about actual food inspection, and risk perception. Even though living in prefectures other than Fukushima was regarded as a protective factor in the first study, living in prefectures other than Fukushima was regarded as a risk factor in the 14th study. Although crisis communication regarding radiation was conducted immediately after the accident, it has been revealed that most of the residents had no knowledge about radiation and were anxious about its potential impact on their lives [11]. As a result, their awareness and refraining from purchasing foods produced in affected areas to avoid radioactive materials increased. However, it is possible that with time the residents of the Fukushima Prefecture have acquired knowledge of radiation, as well as knowledge of radioactive materials in food and food inspections due to radiation risk communication initiatives [12]. On the other hand, it is conceivable that 10 years have passed without residents outside the Fukushima Prefecture gaining sufficient knowledge about radiation. Their awareness and avoidance of Fukushima food products may have become fixed as they may have not updated their knowledge about radiation. In fact, in a survey on the perception of radiation risks conducted among Tokyo residents, about half of the respondents answered that Fukushima residents would experience cancer and other adverse effects in later life, indicating that knowledge is not being disseminated [13]. Furthermore, according to a survey by the Japanese Ministry of Agriculture, Forestry, and Fisheries, 38.6% of people living in urban regions, including Tokyo, reported that they consumed organic vegetables at least once a week, a higher percentage than people living in other areas [14]. Although organic vegetables are generally known to be expensive, people living in urban regions choose

Discussion
In the present study, we investigated factors associated with the public's perception regarding refraining from purchasing foods produced in affected areas to avoid radioactive materials and avoidance of foods produced in the Fukushima Prefecture and clarified the trend over time through 14 surveys.
In the first study, "Refraining from purchasing foods produced in affected areas to avoid radioactive materials" was associated with the following factors: residential area, age, sex, living with infants, knowledge about actual food inspection, and risk perception. On the other hand, in the 14th study, it was associated with the following factors: residential area, marital status, living with elementary school students, knowledge about actual food inspection, and risk perception. Even though living in prefectures other than Fukushima was regarded as a protective factor in the first study, living in prefectures other than Fukushima was regarded as a risk factor in the 14th study. Although crisis communication regarding radiation was conducted immediately after the accident, it has been revealed that most of the residents had no knowledge about radiation and were anxious about its potential impact on their lives [11]. As a result, their awareness and refraining from purchasing foods produced in affected areas to avoid radioactive materials increased. However, it is possible that with time the residents of the Fukushima Prefecture have acquired knowledge of radiation, as well as knowledge of radioactive materials in food and food inspections due to radiation risk communication initiatives [12]. On the other hand, it is conceivable that 10 years have passed without residents outside the Fukushima Prefecture gaining sufficient knowledge about radiation. Their awareness and avoidance of Fukushima food products may have become fixed as they may have not updated their knowledge about radiation. In fact, in a survey on the perception of radiation risks conducted among Tokyo residents, about half of the respondents answered that Fukushima residents would experience cancer and other adverse effects in later life, indicating that knowledge is not being disseminated [13]. Furthermore, according to a survey by the Japanese Ministry of Agriculture, Forestry, and Fisheries, 38.6% of people living in urban regions, including Tokyo, reported that they consumed organic vegetables at least once a week, a higher percentage than people living in other areas [14]. Although organic vegetables are generally known to be expensive, people living in urban regions choose organic foods because they have high levels of consciousness about the food they eat [15]. Therefore, in the present study, residents living in metropolitan areas who avoided foods produced in the Fukushima Prefecture could have avoided foods produced there because of their high consciousness of food safety.
As for the factors age and sex, in the first survey, age groups 30 and older were a risk factor for the avoidance of Fukushima food products. However, in the 14th survey, age was not a risk factor. In addition, being female was a risk factor in the first survey, but no sex association was found in the 14th survey. It has been shown that older age is associated with higher risk perception [16], and the results of the present study were consistent with these findings. Additionally, a previous study reported that females tended to have higher risk perceptions than males [17]. At the time of the first survey, the effects of radiation caused by the Fukushima accident were widely reported on TV and other media, and the accident was a matter of social concern. Therefore, it is possible that this is associated with the attribute of higher risk perception in older people and females. On the other hand, since the 14th survey was conducted 10 years after the accident, we considered that the accident was no longer a big concern, even among those who perceived the risks to be relatively high. That has likely to have contributed to the decrease in the tendency to avoid Fukushima food products.
In the first survey, "living with infants" was associated with refraining from purchasing foods produced in affected areas to avoid radioactive materials, and in the 14th survey, "living with elementary school children" was associated with the same. We considered that this result was related to the fact that the children who were infants at the time of the first survey had become elementary school students by the time of the 14th survey. It is well known that radiation has a significant impact on children, and concerns about pediatric thyroid cancer, which was reported in large numbers after the Chernobyl nuclear accident [18], were particularly high after the Fukushima accident. Therefore, it was assumed that the tendency to refrain from purchasing foods produced in affected areas to avoid radioactive materials was higher among parents who were concerned about potential adverse effects on their children.
In the 14th survey, the experience of bereavement was associated with refraining from purchasing foods produced in affected areas to avoid radioactive materials. It has been shown that spousal bereavement contributes to poor mental health [19]. In addition, the correlation between radiation risk perception and mental health has also been reported [20]. Therefore, this leads to a tendency to refrain from purchasing foods produced in affected areas to avoid radioactive materials and was related to the fact that the experience of bereavement from a spouse caused a decline in mental health and increased radiation risk perception. The fact that those who lost their spouses were not associated with refraining from purchasing foods produced in affected areas to avoid radioactive materials in the first survey is considered to be so because there were relatively more people who avoided Fukushima products several years after the Fukushima accident. Therefore, it is presumed that "those who had experienced bereavement from a spouse," who may have perceived radiation as a high-risk, did not show a significant association.
In the first survey, "knowing about the inspection of radioactive materials in food is implemented in 17 prefectures around the eastern region" was a risk factor for the avoidance of food products made in the Fukushima Prefecture. In the 14th survey, "knowing that if inspection results exceeded screening level, inspection by Ge detector would be implemented" was a risk factor. In previous studies of risk perception, a negative correlation was reported between increased knowledge and decreased risk perception [21], but there was also a positive correlation between increased knowledge and increased risk perception [22]. The results of the present study revealed that gaining knowledge was associated with a tendency to refrain from purchasing foods produced in affected areas to avoid radioactive materials. Notably, the 14th survey revealed that "not knowing that food inspections are being conducted" was a protective factor for avoiding Fukushima food products, suggesting that continued food inspections possibly result in the food products suffering reputational damage.
Furthermore, in both the first and 14th surveys, regarding the radiation risk perception "cannot accept less than the standard value" was a risk factor for refraining from purchasing foods produced in affected areas to avoid radioactive materials. This could indicate that risk perception is not only a judgment based on scientific knowledge but is also an emotional decision.
Looking at the changes in the awareness and avoidance of Fukushima food products over time, it was clear that the percentage of the public avoiding Fukushima food products declined year by year. In Japan, the Reconstruction Agency and other organizations have been implementing measures to reduce reputational damage to Fukushima products [23]. In particular, the Ministry of the Environment has been conducting risk communication activities regarding health effects such as the citizens' round-table conference [24]. In addition, the Food Safety Commission [25], Ministry of Health, Labor, and Welfare [26], Ministry of Agriculture, Forestry and Fishery, and the Consumer Affairs Agency [27] have been conducting risk communication activities regarding food products, and as a result, the avoidance of Fukushima products has reduced. On the other hand, it has been revealed that about 8% of the public still has an aversion to food products made in Fukushima Prefecture. Although food products made in Fukushima have gone through rigorous standard values before being released to the market, some members of the public are judging the risks based on emotions rather than scientific facts. The reason for the continued food inspections is to ensure food safety [28], even though most of the foods have not exceeded the standard values or been detected to have radioactive materials. On the other hand, it was found that knowing the details of food inspection was associated with refraining from purchasing food produced in the affected areas to avoid radioactive materials. Therefore, it is conceivable that food inspections to ensure food safety did not lead to relief among consumers. The results of the present study showed that knowledge of detailed food inspections was associated with refraining from purchasing foods produced in the affected area. Additionally, sex and family members living in the same household were also associated with this, as in our hypothesis. We highlighted the strength of the present study, i.e., the making of comparisons regarding refraining from purchasing foods produced in the affected area between three years after the Fukushima accident and 10 years after that and to confirm differences in the passage of time and associated factors. The findings of the present study are likely to help dispel ongoing reputational damage after the Fukushima accident and prevent this happening to foods during future nuclear disasters that may occur.
This study had several limitations. First, this survey was not conducted for the whole of Japan, so it did not show trends for Japan as a whole. Second, since the present study was a panel survey rather than a survey of all residents, those who were interested in the present study responded to the survey, possibly creating a biased sample.

Conclusions
The tendency to avoid food produced in the Fukushima Prefecture has been decreasing year by year. Ten years after the accident, the tendency to avoid food produced in the Fukushima Prefecture was found to be only about 8%, suggesting that avoidance of Fukushima products in Japan had decreased. In addition, the following factors were associated with refraining from purchasing foods produced in affected areas to avoid radioactive materials in a survey conducted three years after the accident: "living in the Fukushima Prefecture", "being over 30 years old", "female", "living with infants", "knowing the details of food inspection practices" and "not being able to accept radiation risk even if the level is below the standard". On the other hand, 10 years after the accident, the factors "living in the metropolitan area", "having experienced bereavement", "living with elementary school-aged children", "knowing the details of food inspection practices", "knowing that food inspections are conducted" and "not being able to accept radiation risk even if the level is below the standard" were risk factors. Even if the number of years since the accident differed, "living with children", "knowing detailed information about food inspections" and "not being able to accept radiation risk even if the level is below the standard" were commonly associated. Not only did this study reveal that some people's risk perceptions are fixed even when new knowledge is provided, but it also suggests that the implementation of food inspection itself possibly promotes reputational damage. The results of this study may help develop countermeasures against reputational damage to food products after future nuclear disasters.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the Japan Society for the Promotion of Life Sciences (approval number: 2020-5).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: Not applicable.