Associations between bone fractures and post-traumatic stress disorder after the Great East Japan Earthquake in the older adult: a prospective cohort study from the Fukushima Health Management Survey

Background Bone metabolism is known to be affected by psychological stress, and increased psychological stress could increase fracture risk. The purpose of this study was to clarify the effect of mental health deterioration caused by disaster on fracture risk in the older adult.


Background
The Great East Japan Earthquake of magnitude 9.0 occurred on March 11, 2011, with its epicenter at the sea oor 130 km off the southeast Oshika Peninsula, Miyagi Prefecture 1 .The related accident at Fukushima Daiichi Nuclear Power Station (FDNPS) occurred in Fukushima Prefecture.Compared with natural disasters, man-made disasters have been found to be responsible for a higher occurrence of post-traumatic stress disorder (PTSD) 2 .In fact, surveys of evacuation zone inhabitants have indicated that the proportion of adults with a PTSD checklist (PCL) score higher than the cutoff value, re ecting the presence of traumatic symptoms (2011: 21.6%; 2012: 18.3%), was comparable with that of workers affected by the 9/11 World Trade Center attack [3][4] .A survey of 240 evacuees in Hirono Town, Fukushima Prefecture, one of the evacuation areas, found that 66.8% had reported symptoms of depression and 53.5% had reported clinically relevant symptoms of PTSD 5 .Thus, residents in evacuation areas such as those in Fukushima Prefecture could be at high risk of developing PTSD due to events related to the disaster.
A possible association between increased fracture risk and PTSD has recently been reported 6 ; therefore, residents in Fukushima evacuation areas who are at high risk of developing PTSD could also be at high risk of fractures.According to the results of the 2013 Basic Survey on National Life published by the Japanese Ministry of Health, Labor, and Welfare, 11.8% were certi ed as requiring support or nursing care because of falls or broken bones, a major factor equivalent to stroke 7 .In addition, Tanji et al. reported that those with higher psychological distress after an earthquake have a higher risk of requiring nursing care 8 .If residents in evacuation areas in Fukushima Prefecture are at high risk of developing PTSD, the associated higher risk of increased fractures could affect the healthy life expectancy and quality of life (QOL) of similar residents in evacuation areas.In particular, an increase in fracture risk in the older adult would contribute to an increase in the number of those requiring support or nursing care.However, no study has examined the relationship between the risk of developing PTSD due to earthquake and fractures in older adult residents of evacuation areas such as Fukushima Prefecture.Therefore, to maintain and improve the healthy life expectancy and QOL of older adult residents in the Fukushima Prefecture evacuation area, the association between the risk of PTSD and fractures in these residents should be investigated.
This study used data on the older adult taken from the Fukushima Health Management Survey to investigate the association between the frequency of fractures and factors related to the Great East Japan Earthquake, including the risk of developing PTSD.

Study group
Following the Great East Japan Earthquake, the Japanese government declared that the area outside the 20 km radius of the Fukushima Daiichi power plant, where the total radiation dose (cumulative radiation dose) was expected to reach > 20 mSv/year, was the "planned evacuation area"; the area within a 20-to 30 km radius was the "area prepared for emergency evacuation"; and the area within a 20 km radius was the "caution zone."Accordingly, the Japanese government evacuated residents from 13 cities, towns, and villages (all areas within Hirono-machi, Naraha-machi, Tomioka-machi, Kawauchi-mura, Okuma-machi, Futaba-machi, Namie-machi, Katsurao-mura, Iitate-mura, part of Tamura City, part of Minami-soma City, part of Kawamata-machi, and part of Date-city) 9 .
Since January 18, 2012, the Fukushima Health Management Survey has been conducted to evaluate the impact of radiation and to determine the health status of Fukushima residents, considering the diffusion of radioactive substances and evacuation due to the accident at FDNPS, which could subsequently help prevent illnesses, detect illnesses early, and provide early treatment for maintaining and improving the future health of the residents 10 .Among the individuals who completed the Fukushima Health Management Survey, including the Mental Health and Lifestyle Survey component, were those who had been residing in the 13 towns that had to be evacuated because of the Great East Japan Earthquake (individuals with registered residency).Some 180,604 were aged ≥ 15 years (individuals born before April 1, 1995), of whom 27,066 were aged ≥ 65 years (men: 11,795; women: 15,271) and had completed the 2011 edition of the registered questionnaire (valid responses were obtained from 73,431 individuals, with a response rate of 40.7% and a mean age of 56.4 years), which comprised the sample for the present study.According to the questionnaire on "History of fractures after age 50," as of 2011, 19,844 individuals (men: 9,297; women: 10,547; mean age: 75.2 ± 7.0 years) were con rmed to have had no recent history of fractures.Based on a questionnaire on fractures from 2012 to 2016, the occurrence of fractures was determined.Excluding 6,076 individuals whose fracture occurrence could not be tracked until 2016, 13,768 patients (men: 6,632; women: 7,136) were ultimately targeted (with a mean age of 74.5 ± 6.6 years and a mean follow-up period of 3.75 ± 1.48 person-years) (Fig. 1).
The analysis includes age, sex, physical factors (history of fractures, cancer, stroke, heart disease, diabetes, dyslipidemia, hepatic disorder, high blood pressure, and thyroid disease), social factors (experience of earthquake, tsunami, nuclear power plant accident [heard the explosion], need for assistance, change in employment status, and change in residence), psychological factors (history of mental illness, PCL), and lifestyle factors (history of smoking, history of drinking, sleep satisfaction levels, and exercise habits) in the items of the self-administered questionnaire used in the 2011 Mental Health and Lifestyle Survey.

Fracture Determination
In the Mental Health and Lifestyle Survey, questions regarding the presence or absence of fractures differ depending on the year, and the incidence of fractures was determined by combining the questions.
The presence or absence of fractures in 2011 and 2012 was con rmed by responding to a question on "A history of fractures after age 50."In 2013, in addition to the above question, a combination of answers regarding whether "a fracture was diagnosed by a physician within the past year" had been used to determine the presence or absence of fracture.In 2014 and 2015, the presence or absence of fractures had been determined based on only the answer to "fractures within 1 year."In 2016, the incidence of fractures had been determined by a question on "History of fractures after the age of 50."

Trauma Reaction Evaluation
The presence or absence of a trauma reaction was evaluated using PCL [11][12] ; those with a total PCL score of 44 or more were was de ned as being at high risk of PTSD.

History Of Disease
Residents were asked whether they had a history of cancer, stroke, heart disease, hypertension, diabetes, dyslipidemia, hepatic disorder, thyroid disease, or mental illness.

Lifestyle
The questionnaire section on smoking habits comprised three choices: never smoked, former smoker, or current smoker.The section on drinking habits also comprised three choices: never drinks or rarely drinks (less than once a month), former drinker, or current drinker (more than once a month).The question regarding the sleep satisfaction component comprised four choices: satis ed with sleep, slightly unsatis ed with sleep, quite unsatis ed with sleep, and very dissatis ed with sleep or does not sleep at all.Furthermore, the question regarding the exercise habits component comprised four choices: almost daily, two to four times a week, approximately once a week, or almost never.

Experience Of The Great East Japan Earthquake
The question regarding the experience of the Great East Japan Earthquake involved individuals responding to whether they had experienced the earthquake, tsunami, and nuclear power plant accident (heard the explosion).

Need For Assistance
The question regarding need for assistance involved individuals responding to whether they could eat, change clothes, use restrooms, and shop alone.Those who answered that assistance was required for any of the four items were de ned as a group requiring assistance.

Changes In Employment Status
Regarding change in employment status (job change or unemployment) following the earthquake and accident, residents could respond with either "changed" or "unchanged."

Changes In Housing And Evacuation
Residents could respond to the question regarding change in residence after the earthquake in one of the following ways: residing in a shelter, in temporary housing, renting a house or apartment, residing in a relative's house, residing in their own house, or other (free comment).The respondents who had lived in temporary or evacuation shelters immediately after the earthquake were de ned as having changed their residence.
In addition, the residents of Tamura city, Minamisoma-city, Date-city, and Kawamata-towns whose current housing was not a temporary or evacuation center in 2011 were de ned as the nonevacuee group.The others were de ned as evacuee groups.

Statistical Analysis Methods
The statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC, USA).The Kaplan-Meier method and the log-rank test were used to compare the difference in the incidence of fractures based on the answers for each item.In addition, univariate and multivariate Cox proportional hazard models were used to obtain crude and adjusted hazard ratios (HRs) and 95% con dence intervals (CIs) for the association between each factor and fractures.All the data are presented as the number of individuals (n), mean, standard deviation, or percentages.P < 0.05 was considered statistically signi cant.

Participant characteristics
Table 1 lists the participant characteristics.The mean follow-up period for all participants was 3.75 ± 1.48 person-years.Some 1,687 (12.3%) participants experienced a fracture during the observation period.

Survival Analysis Results
The participants were divided into fracture and nonfracture groups, and the relationship between each factor and the incidence of fractures was examined.
As a result of the survival analysis, signi cant differences in fracture incidence in the older adult were found in relation to sex (P < 0.001), PTSD risk (P < 0.001), experience of earthquake (P = 0.013), history of mental illness (P < 0.001), need for assistance (P < 0.001), history of cancer (P < 0.001), history of stroke (P < 0.001), history of heart disease (P < 0.001), history of diabetes (P = 0.015), history of hepatic disorder (P = 0.048), smoking habits (P < 0.001), drinking habits (P < 0.001), sleep satisfaction (P < 0.001), and exercise habits (P = 0.005) (Table 2).The interval scale between the bone fracture and no bone fracture group groups was tested using the log-rank test.

Univariate And Multivariate Cox Proportional Hazard Models
Univariate and multivariate Cox proportional hazard models were performed for each factor, with signi cant differences in the survival analysis to identify the association between psychological indicators and fracture frequency in the older adult (Table 3).

Discussion
Previous studies have reported an increase in the prevalence of diseases such as obesity and lifestyle-related diseases in residents of evacuation areas in Fukushima Prefecture [13][14][15][16] .This increase could be due in part to an increase in stress from changes in the environment caused by moving into temporary housing or living in an evacuation site other than the local area or disturbance in eating habits [17][18][19][20] .Thus, psychological stress is considered to be related to adverse effects on the health of residents, such as those in evacuation areas in Fukushima Prefecture.Those who reported feeling high psychological stress were found to be at increased risk of fracture due to osteoporosis 21 .One possible mechanism for the association between stress and fracture risk is that psychological stress increases cortisol secretion through the hypothalamus-pituitary-adrenal system.Glucocorticoids induce bone loss and increase the risk of osteoporotic fractures 22,23 .Those at high risk of developing PTSD were considered to have high psychological stress immediately after a disaster.Furthermore, older adult people and those with severe living conditions are reportedly more likely to experience worsening symptoms 24 .The psychological effects from the Fukushima nuclear accident have been widespread, causing not only trauma symptoms but also chronic and more complex social problems, such as fragmentation of community and family, and stigma 25 .Therefore, chronic sustained high stress caused by a disaster could contribute to increased fracture risk in the older adult.Thus, to prevent fractures after a disaster, older adult people at increased risk of developing PTSD might need to be assessed for bone mineral density and receive aggressive interventions to reduce psychosocial stress.
Furthermore, it is presumed that those who are at high risk of developing PTSD are more likely to also have other mental disorders, such as depression 26,27 .In fact, residents of Kawauchi Village, one of the 13 municipalities in the evacuation area that exceeded the reference value of Kessler Psychological Distress Scale (K6) scores [28][29][30] were found to have a signi cantly higher percentage of those with PCLs of 44 or higher 31 .In addition, prefectural health surveys have reported that coexistence of PTSD and past mental illness or mental disorders was a poor predictor of mid-term mental health 32 .Thus, the deterioration of mental health caused by a disaster can lead older adult people who already had a tendency to have low physical function in a depressed state to have an even more con ned and sedentary lifestyle, leading to a decrease in physical function and thus increased risk of fractures.Therefore, it is important for older adult people with high PTSD risk and low physical function to maintain and improve their physical function and mental health by encouraging social participation.
Depression itself has also been reported to be associated with an increased risk of fractures 33 .This increase might be mediated by antidepressants 34 .For example, taking one class of antidepressants, selective serotonin reuptake inhibitors (SSRIs), can increase the risk of fractures regardless of depression or bone density 35 .In addition, SSRIs have been reported to contribute to fracture-induced falls and increases in fracture risk 36 .The use of SSRIs is sometimes considered when treating PTSD.Thus, older adult people who are receiving medication for PTSD need to be aware of the risk of fractures due to antidepressants.
We found that those who were very dissatis ed with their sleep were at increased risk of fractures.The prevalence of insomnia and the use of sleeping pills in the Japanese have been reported to increase with age 37 .One of the commonly prescribed sleeping pills in Japan is benzodiazepine or the similar nonbenzodiazepine benzodiazepine receptor agonist.Long-term and high-dose uses of benzodiazepines have been reported to be associated with an increased risk of falls and fractures [38][39][40] .These ndings suggest that pharmacotherapy for insomnia could also be a factor in increasing the risk of fractures in older adult residents, such as those in the evacuation areas in Fukushima Prefecture.The rst choice of treatment for insomnia is to understand the sleep environment and provide sleep hygiene guidance.Our study suggests that securing sleep time and improving the quality of sleep is important to preventing fractures in older adult residents of the evacuation area in Fukushima Prefecture.
Our study found that those with a history of diabetes, heart disease, or stroke were at an increased risk of fractures.Previous studies have reported that the presence of type 2 diabetes is associated with increased fracture risk and paradoxically increases bone mineral density 41 .The incidence of cardiovascular disease has also been reported to be signi cantly associated with the risk of subsequent hip fracture 42 .In addition, stroke has been reported to increase the risk of fractures, including hip fractures, by more than seven-fold within 1 year of hospitalization 43 .Therefore, we anticipate that older adult residents in Fukushima evacuation areas who experienced lifestyle-related diseases and cardiovascular events after the earthquake will likely fall and require nursing care due to the increase in the number of fractures.These ndings suggest that a comprehensive strategy to prevent lifestyle-related diseases and cardiovascular events is necessary to preventing fractures among older adult residents in evacuation areas.
Our study found that women have a higher risk of fractures than men.Contrarily, we found that the risk of developing PTSD was not signi cantly associated with the occurrence of fractures in women.Women often have primary osteoporosis caused by heredity, aging, and postmenopausal decline in female hormones 44 .Patients with osteoporosis have also been found to be more likely to experience a fracture after a fall 45 .Osteoporosis-related fractures can also have a signi cant impact on healthrelated quality of life (HRQOL) 46 .These results suggest that problems characteristic of women, which could not be investigated in this study, could have a greater effect on fracture risk than increased psychological stress.However, exercise can be effective in reducing falls and risk factors associated with fractures from falls in patients with low bone mineral density 47 .Therefore, regular bone density measurements and exercise habit formation for people at high risk of developing PTSD are recommended, especially for women, to prevent fractures from falling and to prevent a decline in HRQOL.
The results of our study revealed that a history of cancer could in uence risk of fracture.Hormonal therapies used in breast and prostate cancer can reduce bone mass and bone density, thus increasing fracture risk [48][49] .Also, individuals with metastatic bone tumors are known to be susceptible to fractures 50 .Thus, older adult patients with cancer are prone to fractures and are very likely to be bedridden if they fall.Therefore, those who have a history of cancer need assistance with creating an environment to prevent falls and to be given guidance on living.
Our study had some limitations.First, the age-adjusted prevalence of post-traumatic stress is known to decrease year by year, and the mental health of residents in evacuation areas in Fukushima Prefecture has improved compared with that at the time of the earthquake 51 .However, it is unclear whether this improvement is re ected in the 13 municipalities' residents, because the ratio of the participants in our study to the total number of participants in the Fukushima residents' health survey is low.Horikoshi et al. had also reported that those who did not respond to the mental survey had a signi cantly higher rate of psychological distress than respondents 52 .Therefore, the results of this study could have underestimated the impact on fractures due to the increased risk of PTSD from the Great East Japan Earthquake and the nuclear accident.Therefore, it might be necessary to survey the mental health of non-respondents by expanding the scope of psychological research and to make efforts to improve it.
Second, in this survey, there were no detailed medication conditions, bone density tests, fracture sites, or questionnaires to survey the situation when a fracture occurs, presence of osteoporosis, or use of antidepressant drugs.As a result, factors that could contribute to fracture risk, such as osteoporosis and the use of antidepressants and steroids 53 , could not be investigated.In addition, the effects of sex differences on fractures could not be completely clari ed because women were not surveyed for menopause or hormone levels.Future studies should include examinations and interviews for these factors.
Finally, studies in postmenopausal women have reported that obesity and underweight are both risk factors for fractures 54 .In this survey, however, there were no survey items on height and weight in FY2011, and Body Mass Index (BMI) could not be calculated.Therefore, in the next study, we plan to examine the relationship between weight and fractures by evaluating health checkup data.

Conclusions
Taken together, the results of our study suggest that chronic psychological stress and reduced sleep time and quality due to disaster could contribute to increased fracture risk.Therefore, understanding bone mineral density, offering active psychological care to reduce psychosocial stress, and providing guidance on sleep are important for preventing fractures in older adult residents, such as those in evacuation areas. Figures

Figure 1 Selection
Figure 1

Table 2
The association between fracture and Mental Health and Lifestyle Survey items

Table 4
The results of multivariate Cox proportional hazard models by sex.
a, b Adjusted for age, PCL score, experience of earthquake, history of mental illness, need for assistance, history of cancer, history of stroke, history of heart disease, history of diabetes mellitus, history of hepatic disorder, smoking habit, drinking habit, level of sleep satisfaction, and exercise habit.95% CI: 95% con dence interval, HR: hazard ratio, Ref: reference, PCL: post-traumatic stress disorder checklist.Cox proportional hazard model; p < 0.05 was considered statistically signi cant.