Introduction
On March 11, 2011, a series of tsunami waves resulting from a 9.0 magnitude earthquake struck the eastern coast of Japan. The subsequent power outage at the Fukushima Daiichi Nuclear Power Plant (FDNPP) led to a shutdown of the cooling system of 3 reactors, resulting in reactor meltdowns and hydrogen explosions from units 1 to 3. 1 The Nuclear and Industrial Safety Agency rated this event as Level 7 (major accident) on the International Nuclear and Radiological Safety Event Scale, 2 which indicates a large-scale release of radioactive material with widespread health and environmental effects, that requires implementation of planned, and extended countermeasures. 3 Accordingly, the Governor of Fukushima issued an evacuation decree at 8.50PM on the same day for all persons residing within 2 km of the FDNPP, which the Prime Minister of Japan then extended at 9.23PM to all residents within 3 km. As a determinative step, the evacuation order was expanded the following day to all residents living within 20 km of the power plant. This included the town of Tomioka, 4 which is located in the 10 – 20 km belt around the FDNPP.
Risk perception is defined as the evaluation of hazards through intuitive judgement by the public.
Based on the collection of diverse information sources and the codification of past experiences, risk perception allows for a transformation of behaviors and the surrounding environment, resulting in either risk reduction or creation. Reference Slovic5 Nuclear power, with its unobservable, novel characteristics, and inherent capacity to cause delayed health consequences, is judged by the general public as harboring uncontrollable risks, possessing catastrophic potential, fatal consequences, and bearing an inequitable distribution of risks and benefits. The combination of the country’s past experiences with atomic bombings, the complex underlying mechanisms of nuclear accidents, the accompanying delayed stochastic health effects of radiation exposure led to widespread anxiety, and dread risk among Japanese citizens following the FDNPP disaster. Reference Nakayachi, Yokoyama and Oki6 Although the level of radiation exposure estimated to be received in evacuees of Fukushima prefecture 4 months after the accident was limited, and no appreciable radiation-related health effects were expected, Reference Ishikawa, Yasumura and Ozasa7,8 risk perception regarding consumption of food and tap water, effects on general health, and genetic effects on future generations remained a concern up to 6 years later. Reference Yasumura, Hosoya and Yamashita9–Reference Maeda and Oe11 It has been established that sociocultural background and immediate response to trauma are also fundamental drivers behind risk perception among the general public. Reference Slovic5,Reference Takebayashi, Lyamzina, Suzuki and Murakami12–Reference Slovic and Peters14 However, excessive degrees of risk perception have been linked to poor health through increased levels of psychological distress, Reference Suzuki, Yabe and Yasumura15,Reference Bromet16 adverse mental health status, Reference Miura, Nagai and Maeda17 reduced frequency of laughter, Reference Murakami, Hirosaki and Suzuki18 and incitation of feelings of social unrest through behaviors promoting discrimination against refugees, Reference Sills, Murakami, Kumagai, Stojarov and Tsubokura19,Reference Sawano, Nishikawa, Ozaki, Leppold and Tsubokura20 as well as avoidance of food products from affected areas, Reference Yoshizawa21 and exaggerated notions of the dangers of the surrounding environment. Reference Suzuki, Murakami, Nishikiori and Harada22 Additionally, risk perception and health status dictate decision-making such as the intention to return (ITR) to hometowns by evacuees. Reference Slovic and Peters14
Previous studies have demonstrated an inverse correlation between high degrees of radiation risk perception and the desire to return to Tomioka and Okuma towns after the FDNPP disaster. Reference Matsunaga, Orita and Iyama23,Reference Orita, Hayashida, Urata, Shinkawa, Endo and Takamura24 We consider the present study unique in that it evaluates the change in risk perception and the resulting relationship to the affected populations’ intentions to return over time (between 2017 and 2021) in a specific town, unlike the study of Hagen et al. Reference Hagen, Opejin and Pijawka25 Several studies have measured statistical change in risk perception among a subpopulation residing in a contaminated area who were exposed to the nuclear accident and to continuous and targeted risk communication strategies during the recovery phase of the disaster. However, unlike the studies of Suzuki et al., Reference Suzuki, Takebayashi and Yasumura26 Ito et al., Reference Ito, Goto and Ishii27 and Kohzaki et al., Reference Kohzaki, Ootsuyama, Moritake, Abe, Kubo and Okazaki28 the present study was conducted 6 and 10 years after the disaster rather than in its immediate aftermath. Thus, the results are more indicative of responders’ longer-term behaviors and views, and of what factors, if any, play a role in the change of perception of risk and ITR among this specific subpopulation.
All residents of Tomioka were evacuated until April 2017, when the Japanese government lifted the evacuation order, barring the ‘difficult-to-return’ zones of Tomioka. Recovery efforts, including decontamination of environmental radionuclides and reconstruction of the town infrastructure, were spearheaded by the Tomioka town office in the interest of facilitating the return of its residents. Based on published evidence of reduced radiation-related anxiety and fear among participants in small-group consultations and explanations/ seminars with scientific experts, Reference Hino, Murakami and Midorikawa29–Reference Midorikawa, Tanigawa, Suzuki and Ohtsuru31 similar community recovery efforts were organized by involved stakeholders at Tomioka town, and comprised of crisis and risk communication sessions, regular publication of periodicals, the establishment of a satellite research office, environmental monitoring, a ‘food inspection center’ that allowed residents to measure radionuclide levels in locally sourced produce, and stationing of a public health nurse for individual consultations. Reference Takamura, Orita, Taira, Matsunaga and Yamashita32 The pace of population growth ensured that Tomioka town would fulfil its hope of increasing its population to at least 2000 by March 2022, 5 years after the evacuation order was lifted. However, as of 2021, the number of former residents among this new population remained limited at 11%. This is in contrast to the 63% return migration that occurred within 13 months, in 12 counties located within 100 miles of the path of Hurricane Katrina, after an 85% evacuation rate. Reference Groen and Polivka33 Similarly, after forced evacuation resulting from the 2010 Mt. Merapi earthquake in Indonesia, the return rate was 52% among survey respondents, and another 17% were in transition to return 16 months after the disaster. Reference Muir, Cope, Angeningsih, Jackson and Brown34 Since 2017, questionnaires that measure risk perception and evacuees’ ITR have been distributed to current residents of Tomioka and to the evacuation addresses of current evacuees (who are still registered with the Tomioka town council). We considered that it was necessary to review the effectiveness of ongoing risk communication activities and to identify potential areas for improvement as the population of the town increased. The present study aims to examine the change in risk perception and ITR between 2017 and 2021. The results of this study will guide risk communication strategies and the long-term community recovery effort in Tomioka town.
Methods
Participants
The assessment in this study is based on survey responses that were collected from residents and evacuees from Tomioka town in September, 2017. Reference Matsunaga, Orita and Iyama23 Additionally, in November 2021, we distributed questionnaires to all residents and evacuees who were able to receive mail from the municipal office. The study participants were all residents and evacuees of Tomioka town aged ≥ 18 years who held resident cards as of October 2021. (Supplementary Table 1. Current location of evacuees by February 1, 2023)
Ethical issues
The basis and purpose of the study were explained in a letter attached to the questionnaire, along with a privacy notice. Consent to participate was voluntary, and it was explained that participants who choose not to participate would face no disadvantages. Written informed consent was obtained from all participants through the return postage of the questionnaire. Participation was anonymous and individuals could not be identified. No minors participated in the study. All study protocols were approved by the ethics committee of Nagasaki University Graduate School of Biomedical Sciences (Approval No. 21082702, September 6, 2021).
Questionnaire
The questionnaire used in the present study was based on that developed and distributed among Tomioka residents and evacuees in 2017. Reference Matsunaga, Orita and Iyama23 Questions included demographic information (age, sex, currently living with a child or grandchild < 18 years old or not), ITR to Tomioka, and desire to consult radiation experts. We asked residents whether they were reluctant to eat food produced and collected in Tomioka, and to drink tap water from Tomioka. We also asked questions to evaluate subjective risk perception about the potential health effects of radiation exposure, such as the risk of cancer to themselves, and genetic effects in the next generation caused by living in Tomioka. Responses were in the form of yes/ no or multiple choice as appropriate. Questions regarding ITR and risk perception were scaled using a 4-point scale (1 = Strong agreement, 2 = Probably yes/ A lot, 3 = Probably no/ A little, 4 = Strong disagreement). We chose to capture risk perception and related behaviors through Lindell’s 4-point Likert scale. Reference Lindell and Barnes35
Statistical analysis
The present study analyzed the change in risk perception and ITR among Tomioka residents and evacuees, between the responses received in 2017 and those received in 2021. Responses were set as user missing if they were incomplete or inconsistent (if multiple responses were inappropriately selected), and we conducted pairwise exclusion of missing values. Responders were divided into 2 cohorts according to the year of response. ‘Age’ was converted from the decade of life to young (< 60 years) and elderly (≥ 60 years) residents. ‘ITR’ was converted from the 4-point scale into 3 categories as follows: ITR (+) (already returned or wanted to return), unsure (about returning), and ITR (–) (did not want to return). ‘Risk perception’ was similarly converted from the 4-point scale to either yes or no.
Demographic characteristics were compared between the groups using the chi-square test and missing data were noted. Variables significantly associated with each of the ITR (+), unsure, and (–) categories were then identified in both groups. Significant factors were analyzed using logistic regression, and odds ratios with 95% confidence intervals (95% CI) were obtained. Data analysis was performed using IBM SPSS Statistics version 28 (IBM Corp., Armonk, New York, USA). P-values < 0.05 were considered statistically significant.
Results
Demographic characteristics
The response rate was 27.3% (2269 responses) in the 2017 survey, Reference Matsunaga, Orita and Iyama23 and 34.0% (2899 responses) in the present 2021 survey.
The proportion of elderly responders was 70.8% in 2021 and 63.3% in 2017, Reference Matsunaga, Orita and Iyama23 (P < 0.001). Most responders were not living with children or grandchildren under the age of 18 years at the time of the survey (80.0% and 84.5% in 2017 and 2021, Reference Matsunaga, Orita and Iyama23 respectively, P < 0.001). The groups did not differ significantly based on sex (Table 1. Demographic characteristics, ITR, and risk perception).
Demographic characteristics, ITR, and risk perception
1 N (%)
2 Matsunaga H, Orita M, Iyama K, et al. Intention to return to the town of Tomioka in residents 7 years after the accident at Fukushima Daiichi nuclear power station: a cross-sectional study. J Radiat Res. 2019;60(1):51-8.
* = P -value < 0.05
ITR
In 2017, 8.7% of responders were categorized as ITR (+), 34.4% as ITR unsure, and 57.0% as ITR (–). Reference Matsunaga, Orita and Iyama23 In 2021, these proportions changed to 11.9%, 24.3%, and 63.8% for ITR (+), ITR unsure, and ITR (–), respectively (P < 0.001).
Risk perception and desire to consult radiation experts
In 2017, 80.3% of responders stated that they had anxiety regarding the consumption of food produced in Tomioka, Reference Matsunaga, Orita and Iyama23 whereas only 49.4% of responders reported this feeling in 2021 (P < 0.001). Concerning anxiety about health effects resulting from exposure to radiation in Tomioka, 67.9% reported feeling this anxiety in 2017, Reference Matsunaga, Orita and Iyama23 and 52.5% in 2021 (P < 0.001). Anxiety regarding the genetic effects of radiation exposure in Tomioka town was experienced by 72.5% of residents in 2017, Reference Matsunaga, Orita and Iyama23 and 48.1% in 2021 (P < 0.001). In 2017, 72.9% of residents did not have the desire to consult radiation experts, Reference Matsunaga, Orita and Iyama23 which increased to 88.7% in 2021 (P < 0.001).
Variables associated with ITR
In 2017, the demographic variables of sex, age, and living with children, as well as the desire to consult with radiation experts were significantly associated with ITR. Reference Matsunaga, Orita and Iyama23 Relative to the ITR unsure and ITR (–) groups, the ITR (+) group consisted mostly of males (55.8% vs. 52.6% and 46.2%, P = 0.004), the elderly (74.6% vs. 66.9% and 59.54%, P < 0.001), and responders currently not living with children (90.9% vs. 80.3% and 78.1%, P < 0.001). The desire to consult radiation experts was 35.8%, 36.5%, and 20.2% among the ITR (+), ITR unsure, and ITR (–) groups, respectively (P < 0.001) (Table 2. Variables associated with ITR in 2017 and 2022).
Variables associated with ITR in 2017 and 2022
1 N (%)
2 Matsunaga H, Orita M, Iyama K, et al. Intention to return to the town of Tomioka in residents 7 years after the accident at Fukushima Daiichi nuclear power station: a cross-sectional study. J Radiat Res. 2019;60(1):51-58.
* = P - value < 0.05
In 2021, the demographic variables of sex and desire to consult with radiation experts were significantly associated with ITR. The ITR (–) group consisted mostly of females (55.3%), whereas more male responders were either ITR (+) (52.2%) or ITR unsure (51.3%) (P = 0.005). The proportions of responders who had the desire to consult with radiation experts were 15.2%, 15.4%, and 8.6% in the ITR (+), ITR unsure, and ITR (–) groups respectively (P < 0.001).
Proportionally, elderly responders in the ITR (+), ITR unsure, and ITR (–) groups were 74.7%, 68.5%, and 69.8% respectively, and most responders (81.7%, 83.8%, and 83.7%) who were ITR (+), ITR unsure, and ITR (–) respectively, were not living with children at the time, but these variables were not significantly associated with ITR in 2021 (Table 2. Variables associated with ITR in 2017 and 2022).
Risk perception and ITR
In 2017, the proportions of ITR (+), ITR unsure, and ITR (–) responders who reported having anxiety about the consumption of food produced in Tomioka were 61.0%, 79.5%, and 84.0%, respectively (P < 0.001). Reference Matsunaga, Orita and Iyama23 In 2021, these proportions were 45.2%, 55.3%, and 52.4% (P = 0.018). Risk perception for food was significantly associated with ITR in 2017 and 2021.
Regarding risk perception for self-health, 36.4%, 63.8%, and 75.3% of ITR (+), ITR unsure, and ITR (–) residents in 2017 reported feeling anxiety regarding the effects of radiation exposure in Tomioka town on their health (P < 0.001). Reference Matsunaga, Orita and Iyama23 In 2021, these proportions were 51.0%, 57.3%, and 52.6%. Risk perception about health was significantly associated with ITR only in 2017.
In 2017, the proportions of ITR (+), ITR unsure, and ITR (–) responders who reported feeling anxiety regarding the genetic effects of radiation exposure in Tomioka were 46.7%, 71.5%, and 77.0% respectively (P < 0.001). Reference Matsunaga, Orita and Iyama23 In 2021, these proportions were 44.6%, 52.2%, and 47.9%. Risk perception regarding genetic effects was significantly associated with ITR only in 2017.
Logistic regression
The first model of logistic regression analysis revealed that anxiety regarding consumption of food produced in Tomioka was independently associated with ITR in both 2017 and 2021. Compared to ITR (+) residents, ITR unsure responders had 2.5 higher odds ratio (95% CI, 1.7 - 3.6) in 2017 of having anxiety regarding consumption of food produced in Tomioka and 1.5 times higher odds ratio (95% CI, 1.1 - 2.0) in 2021. ITR (–) responders had 4 times higher odds ratio (95% CI, 2.8 - 5.7) in 2017 of having anxiety regarding consumption of food produced in Tomioka, whereas the odds ratio was 1.4 times higher (95% CI, 1.1 - 1.8) in 2021 (Table 3). Younger ages had 1.8 times higher odds (95% CI, 1.3 - 2.6) of being associated with ITR (–) in 2017, whereas age did not remain significant in 2021.
Logistic regression models
Reference group is ITR+
1 Matsunaga H, Orita M, Iyama K, et al. Intention to return to the town of Tomioka in residents 7 years after the accident at Fukushima Daiichi nuclear power station: a cross-sectional study. J Radiat Res. 2019;60(1):51-8.
* P - value < 0.05
The second model revealed that compared to ITR (+) responders, those who were ITR unsure had 3.4 times higher odds ratio (95% CI, 2.3 - 4.9) of having anxiety regarding health effects on themselves, those who were ITR (–) had 7.3 times higher odds ratio (95% CI, 5.1 - 10.5) in 2017, and anxiety about health effects of radiation exposure was independently associated with ITR in 2017. Nevertheless, in 2021, anxiety about health effects of radiation exposure was not associated with ITR.
The third model of logistic regression analysis revealed that in 2017, responders’ anxiety about genetic effects on the next generation was independently associated with ITR in both the ITR unsure and ITR (–) groups compared to the ITR + group. In 2021 however, anxiety about genetic effects was independently associated with ITR only in the ITR unsure group. All models of logistic regression revealed that the desire to consult radiation experts remained significantly lower among ITR (–) responders in 2017 and in 2021 (Table 3. Logistic regression models).
Limitations
Limitations of the present study include its cross-sectional design, which prevented us from deriving any causal relationship between risk perception and ITR over time. Due to the nature of the disaster, since 2011, various stakeholders have been involved in the rehabilitation of Fukushima. Thus, residents and evacuees have been subjected to multiple rounds of surveys, evoking feelings of ‘survey fatigue’ with time, which is a possible contributing factor towards the limited response rate for our voluntary questionnaire. As a measure to counteract this low response rate and increase the sample size, we mailed our surveys to all households within Tomioka and also to secondary evacuation addresses outside the town among registered evacuees. Reference Kelley, Clark, Brown and Sitzia36 However, surveys conducted in surrounding municipalities have displayed similar response rates. Reference Kuroda, Iwasa, Orui, Moriyama, Nakayama and Yasumura13,Reference Matsunaga, Orita and Iyama23,Reference Orita, Hayashida and Nakayama37–39 There is also an inherent selection bias in those who decided to respond to the survey. The responses were more reflective of responders aged 60 years or older (63% in the 2017 survey and 71% in the 2021 survey) than a reflection of all Tomioka residents. As of 2021, the actual population aged > 65 years in Tomioka town was 32%. 40 It has been hypothesized that as the elderly have stronger feelings of attachment to their hometowns, they are more likely to be involved in rehabilitation efforts such as surveys, compared to the younger population, who are more likely to migrate and establish community ties elsewhere. Reference Hashimoto41 In addition, the responders were heterogeneous between the time points, which made it difficult to determine the long-term effectiveness of risk communication strategies. There is a need to also explore other factors relevant to risk perception among evacuees and residents after a disaster. Regarding face and content validity of our questionnaire, the included questions underwent multiple rounds of evaluation by 2 independent experts in the field of radiation risk communication, who checked whether the questions satisfactorily covered all aspects of the constructs being measured. The questionnaires were also vetted by the local town council regarding the appropriateness of the questions and the adequacy of the instructions in the context of survey responder ability. All experts were native speakers and fluent in the local language. Minor corrections and fine-tuning of the questionnaire were done according to comments and suggestions received. This questionnaire was constructed based on previous published research carried out on a similar sample, Reference Matsunaga, Orita and Iyama23 thus not necessitating a pilot study. However, we could not perform construct validity analysis on this questionnaire, and thus the magnitude of the role of the underlying latent construct is not completely known, which could have resulted in information bias. To reduce this bias, we framed our questionnaire as closely as possible to similar previously published surveys that have measured risk perception and driven behavior in similar samples. Reference Matsunaga, Orita and Iyama23,Reference Matsunaga, Orita, Liu, Kashiwazaki, Taira and Takamura38,Reference Matsunaga, Orita, Oishi, Taira and Takamura42 Although there is no singular or standardized tool to measure risk perception, we chose Lindell’s 4-point Likert scale for its generalizability, comparability across responders, its validation in our target sample specifically, Reference Lindell and Barnes35,Reference Kashiwazaki, Takebayashi and Murakami43 and based on its use in the government-led Fukushima Health Management Survey. Reference Suzuki, Yabe and Yasumura15
Discussion
The present study examined the changes in risk perception and ITR between 2017 and 2021 among residents and evacuees of Tomioka town after the FDNPP disaster. From 2017 to 2021, the magnitude of risk perception reduced significantly for food (from 80.3% to 49.4%), health effects (from 67.9% to 52.5%), and genetic effects (from 72.5% to 48.1%). In determining ITR in 2021, logistic regression models revealed that anxiety for health effects was no longer significant, whereas anxiety for genetic effects remained significant among ITR unsure responders, and food anxiety remained significant among ITR unsure and ITR (–) responders.
From 2017 to 2021, although the odds ratio of risk perception for consumption of food produced in Tomioka reduced from 2.5 to 1.5 among ITR unsure responders, and from 4.0 to 1.4 among ITR (–) responders, its significance in deciding ITR persisted. Environmental studies have revealed that radiocesium (134Cs,137Cs) was the primary radionuclide contaminant among the land surface and crops in the aftermath of the FDNPP accident. Reference Murakami and Oki44 Measured in Fukushima city (around 50 km from the nuclear power plant) in March 2012, the effective doses of134Cs and137Cs were 0.44 µSv/ month among vegetables bought from the market and 2.60 µSv/ month among locally-grown vegetables, which conferred minor ingestion-related lifetime-attributable risks for all solid cancers. Reference Murakami and Oki44 The same study reported effective134Cs and137Cs doses of 0.19 µSv/ month in an uncontaminated area (Tokyo, approximately 230 km from the power plant) measured during the same time period, which resulted from other natural radionuclides in the diet; e.g., potassium 40 (40K) and polonium 210 (210Po). Similarly, in Tomioka in 2019, radiocesium concentrations measured from local produce were much lower than the standard or public dose limits. Reference Yamaguchi, Taira, Matsuo, Orita, Yamada and Takamura45 In the same year, despite relatively high external median air dose levels in the difficult-to-return zones, radionuclide concentrations were low in indoor and outdoor air, and in surface soil in areas in which the evacuation order had been lifted. Reference Matsuo, Taira and Orita46 Nevertheless, residents persistently reported anxiety about consumption of locally produced food; while Japanese citizens (especially those living outside Fukushima), living with children, cognizant of food inspections, and aware of inspection specifics, remained concerned about radioactivity in vegetables and were averse to purchasing Fukushima-produced food products. Reference Yoshizawa21,Reference Matsunaga, Orita and Iyama23,Reference Orita, Mori, Taira, Yamada, Maeda and Takamura47 11 years after the FDNPP accident, these residents are unaccepting of the low radiation risk and still perceive purchasing Fukushima-produced food products as dangerous. Reference Yamaguchi, Horiguchi and Kunugita48
To alleviate this anxiety among its residents, the Tomioka town office established a ‘food inspection center’ that allowed residents to measure radionuclide levels in locally sourced produce. A study that examined the effect of implemented radiological countermeasures on subjective well-being and anxiety of Fukushima residents 5 years after the FDNPP disaster, found that although the food inspection facility had low participation and utilization rates compared with the remaining dose monitoring countermeasures, 79% of participants deemed the facility as highly useful. However, the study also reported a deterioration in self-rated health among these participants, which was not observable for other countermeasures. Reference Murakami, Takebayashi and Takeda49 To allow for the complete rehabilitation of residents so they can live their daily lives in peace and free of radiation-related anxiety, the cultivation of a practical radiological culture is essential. Reference Wada, Nakata and Fukumoto50 The latter is defined by the International Commission on Radiological Protection (ICRP) as the provision of resources aimed at improving ‘the knowledge and skills enabling citizens to make well-informed choices and behave wisely in situations involving potential or actual exposures to ionizing radiation.’ 51 The food inspection center is an example of a space run by experts where residents can conduct dose measurements, read, understand the results, and then arrive at an independent decision regarding their risk perception. Collectively, it is imperative to reinforce the message of safety in the consumption of Fukushima-produced food products for a re-invigoration of the local economy through welcoming of residents and visitors.
Regarding anxiety about genetic effects, although the proportion of residents expressing this anxiety reduced by 24.4% from 2017 to 2021, 48.1% of the 2021 cohort persistently reported high-risk perceptions of genetic effects. Risk perception for genetic effects remained a significant factor in determining ITR among those unsure about returning in 2017 and 2021. In 2017, 74.9% of ITR unsure residents who were living with children had anxiety regarding the genetic effects of future generations (P = 0.028), but there was no significant association of living with children in 2021, indicating either a possible generalization of risk perception amongst all residents or a reduction in the magnitude of anxiety among those living with children. Tallying with our results, the 2018 Fukushima health survey revealed that 53% of evacuees consistently reported high-risk perceptions of the genetic effects of radiation, Reference Suzuki, Takebayashi and Yasumura26 and in Kawauchi village, where return rates were 80% in 2017, 46.1% of residents in 2021 maintained a high-risk perception of genetic effects. Reference Matsunaga, Orita and Iyama23,Reference Liu, Matsunaga, Orita, Taira and Takamura52 In its 2020/ 2021 report, the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) determined that there was no evidence for the occurrence of genetic effects as a result of the doses exposed by Fukushima residents during the FDNPP accident. 53 The average external radiation dose among Tomioka residents was found to be 0.5 mSv per 4 months, 39 as measured by the Fukushima Health Management survey between March and July 2011; however, risk perception regarding radiation effects remained high among residents 7 years later. Reference Matsunaga, Orita and Iyama23 Therefore, it is evident that the formation of risk perception is dependent on alternative, extraneous factors apart from presented scientific data, and this anxiety about genetic effects among Tomioka residents must continue to be monitored.
Concurrent with Murakami et al., Reference Murakami, Takebayashi and Harigane54 the factors of age, sex, and living with children did not play a role in determining ITR in 2021, and the influence of risk perception about health and genetic effects in deciding ITR diminished in significance over time. The number of residents who wished to consult with radiation experts decreased over time for all groups. Although the desire to consult with radiation experts remained significant for determining ITR in the present study for both years, the proportion of residents who responded affirmatively to this question reduced over time, indicating a reduction in risk perception. However, various studies have established that an individual’s perception of risk is dependent on additional diverse factors such as past experiences, reaction to trauma, and evacuation status, as well as recent bereavement, Reference Suzuki, Takebayashi and Yasumura26,Reference Ferrer, Klein, Persoskie, Avishai-Yitshak and Sheeran55 mental health status, Reference Suzuki, Takebayashi and Yasumura26 psychological distress, Reference Suzuki, Yabe and Yasumura15,Reference Miura, Nagai and Maeda17,Reference Kashiwazaki, Takebayashi and Murakami56 and age. Reference Matsunaga, Orita and Iyama23 Other factors include sex, Reference Matsunaga, Orita and Iyama23,Reference Oishi, Orita, Taira, Kashiwazaki, Matsunaga and Takamura57 employment status, Reference Suzuki, Takebayashi and Yasumura26 socioeconomic status, and educational attainment. Reference Suzuki, Takebayashi and Yasumura26,Reference Murakami, Nakatani and Oki58 Our results demonstrated the presence of radiation-related anxiety in half of the 2021 cohort, indicating the chronic and unpredictable nature of risk perception, and its related health consequences. 19 years after Chernobyl, evacuee mothers reported impaired well-being, elevated risk perceptions, and poor mental health in the preceding year due to the enduring impacts of the disaster on their lives. Reference Adams, Guey, Gluzman and Bromet59 These perceptions were retained in their children, who reported similar poor health compared to controls, Reference Bromet, Taormina and Guey60 implying that among survivors, disaster-related high-risk perceptions and mental health effects are lingering and persistent. These disaster-related effects might manifest in the form of safety behaviors, including intolerance of uncertainty and increased perceptions of anxiety through paying selective attention to risks. Safety behaviors (including consulting radiation experts and measuring radioactivity levels of food) are actions that involve the seeking of reassurance against anxiety and are performed to evade emotional distress.
Disaster-related risk communication strategies must evolve with time and should be of a multi-pronged nature. Information providers must first assess and improve residents’ health literacy and trust amongst the local authorities before disseminating scientific material. More so, it has become evident that resident involvement in all aspects of decision making is necessary for community recovery. Reference Figueroa61 It is crucial for stakeholders to conduct targeted and customized risk communication for each demographic of the population, Reference Yamaguchi and Horiguchi62,Reference Yashima and Chida63 and improve the methods of communication strategies with holistic measures of familiarity and ingenuity, based on residents’ needs and concerns. Reference Murakami, Sato and Matsui64 There is an urgent need to formulate a systematic method for capturing risk perception and its change over time. Ultimately, it is crucial to respect residents’ ITR wishes, as the decision is representative not solely of risk perception but of multiple factors such as the social network, town infrastructure, and employment opportunities, among others. Reference Kuroda, Iwasa, Orui, Moriyama, Nakayama and Yasumura13,65
Conclusion
Risk perception for food and genetic effects remains a factor in deciding ITR among Tomioka residents. There is a need for continual risk communication through promoting health literacy among evacuees and improving trust in experts through targeted and multi-pronged strategies.
Supplementary materials
To view supplementary material for this article, please visit https://doi.org/10.1017/dmp.2023.58
Acknowledgement
We would like to thank all study participants and staff members of the municipal government of Tomioka.
Author contributions
Conceptualization: VH and NT; Methodology: MO; Software: MO; Validation: VH, MO and HM; Formal analysis: VH; Investigation: MO and YK; Resources: MO; Data curation: VH; Writing - original draft preparation: VH and MO; Writing - review and editing: YT and NT; Visualization: MO; Supervision: NT; Project administration: NT; Funding acquisition: NT. All authors have read and agreed to the published version of the manuscript
Funding information
This work was supported by the Research Project on the Health Effects of Radiation organized by the Ministry of the Environment, Japan.
Competing interest
The authors declare no conflicts of interest. The funders had no role in the design of the study, in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
Ethical standard
The study was conducted in accordance with the Declaration of Helsinki and approved by the ethics committee of Nagasaki University Graduate School of Biomedical Sciences (approval No. 21082702, 6 September 2021). Informed consent was obtained from all subjects involved in the study
Open access
