Psychological Recovery 5 Years After the 2004 Niigata-Chuetsu Earthquake in Yamakoshi, Japan

Background The 2004 Niigata-Chuetsu earthquake of Japan caused considerable damage. We assessed long-term changes in psychological distress among earthquake victims during the period 5 years after the earthquake. Methods The participants were people aged 18 years or older living in Yamakoshi, a community in Niigata Prefecture near the epicenter. A self-administered questionnaire survey was conducted annually for 5 consecutive years after the earthquake. Response rates were 1316/1841 (71.5%) in 2005, 667/1381 (48.3%) in 2006, 753/1451 (51.9%) in 2007, 541/1243 (43.5%) in 2008, and 814/1158 (70.3%) in 2009. The questionnaire asked about demographic characteristics, including sex, age, employment status, social network, and psychological status. Psychological distress was assessed using the 12-item General Health Questionnaire and was defined as a total score of 4 or higher. Results The overall prevalence of psychological distress decreased (P < 0.0001) gradually from 2005 (51.0%) to 2008 (30.1%) but tended to increase from 2008 to 2009 (P = 0.1590). Subgroup analyses showed that prevalence did not decrease over the 5-year study period among participants with poor social contact (P = 0.0659). From 2008 to 2009 prevalence increased in women (+7.5%, P = 0.0403) and participants aged 65 years or older (+7.2%, P = 0.0400). Conclusions The prevalence of psychological distress in Yamakoshi people decreased steadily during the 4 years immediately after the earthquake but appeared to increase thereafter. The earthquake victims are still reestablishing their lives. Thus, continued attention should be focused on maintaining and further assessing their mental health.


INTRODUCTION
The 2004 Niigata-Chuetsu earthquake in Japan had a nearepicenter maximum seismic intensity of 7 on the Japan Meteorological Agency's Intensity (JMAI) scale and killed more than 60 people. Approximately 103 000 people sought refuge, and 16 000 houses were destroyed. 1 A unique characteristic of the Niigata-Chuetsu earthquake was that mountainous areas experienced considerable damage on the ground and disruption of both traffic and communication networks. 2 Such a large earthquake adversely affects the psychological status of people. After the Niigata-Chuetsu earthquake, various short-term psychiatric problems and symptoms of psychological distress were documented among the victims. 3 Psychological problems appear to persist long after earthquakes. Previous epidemiologic studies found that the prevalence of psychological distress remains elevated many years after earthquakes. 4,5 Similar long-standing adverse effects on psychological health may have been present after the Niigata-Chuetsu earthquake. Thousands of people in the disaster area required temporary housing for 3 years, 6 and financial problems due to property damage and/or unemployment after the earthquake were long-lasting. 2 Thus, the long-term effects of psychological distress after the Niigata-Chuetsu earthquake warrant investigation.
The village of Yamakoshi (currently part of Nagaoka City) was located near the epicenter of the earthquake, where physical damage was enormous. All village residents sought refuge in a nearby city after the event. To care for these people, annual health checks were conducted for 5 consecutive years after the earthquake. In this setting, the psychological status of Yamakoshi residents was also evaluated. The aim of this study was to assess long-term changes in psychological distress among the resident population during the 5 years after the Niigata-Chuetsu earthquake.

METHODS Participants
The Niigata-Chuetsu earthquake occurred on October  This subgroup was also analyzed for comparison with the results of whole-population analyses.

Procedure
Self-administered questionnaires for the years 2005 through 2008 were delivered and collected by hand (for participants living in temporary refugee housing) or sent and collected by mail (for other participants). The 2009 questionnaires were collected by hand from all participants. The questionnaires elicited demographic information on sex, age, employment status, social network, and psychological status. Regarding social networks, participants were asked, "Do you easily talk with or call on neighbors in the community?". Answers were categorized as "often," "sometimes," "not often," or "rarely." Participant psychological status was assessed using the 12item General Health Questionnaire (GHQ-12). This study received approval from the Ethics Committee of the Niigata University School of Medicine.

GHQ-12
The outcome of this study was level of psychological distress, as measured by the previously validated Japanese version of the GHQ-12. 8,9 The GHQ is a self-administered tool originally developed as a screening tool for non-psychotic psychiatric illnesses 10 and is used in clinical settings and in epidemiologic studies of mental health status in general populations.
Participants are asked to assess their moods, feelings, and behaviors during the previous 4 weeks, and the tool is used to evaluate their responses on a 4-point response scale. All GHQ-12 items were rated using the GHQ method (0-0-1-1) before being summed to a total score (range, 0-12). A person with a high GHQ score is considered to have a high level of psychological distress, including depression, anxiety, and other psychiatric symptoms. Honda et al reported that a score of 4 or higher had satisfactory sensitivity and specificity as a GHQ-12 cutoff score in the Japanese general population; we therefore used this value. 9

Statistical analysis
The study outcome was prevalence of psychological distress, as defined by a GHQ-12 score of 4 or higher. The secular trend in the prevalence of psychological distress during the 5year study period was observed and tested using the Cochran-Armitage test for trend. Secular trends were investigated by sex and age group, eg, elderly people (≥65 years of age) versus younger people (≤64 years of age). Secular trends were also investigated according to job status (unemployed vs other) and level of social contact (often/sometimes vs not often/rarely) for the respective years, because lack of employment and loss of contact in the community were reported to be strongly associated with psychological distress in a cross-sectional study of the Niigata-Chuetsu earthquake. 11 The χ 2 test was used to compare prevalence rates between 2 groups. Multiple logistic regression analysis was used to determine if sex (1, female; 0, male), age (1, ≥65 years; 0, ≤64 years), employment status (1, unemployed; 0, other), and social contact (1, not often/rarely; 0, often/sometimes) were independently related to the outcome variable. Data were analyzed utilizing SAS statistical software (release 9.1.3, SAS Institute Inc., Cary, NC, USA). A P value of less than 0.05 was considered to indicate statistical significance. Table 1  Secular changes in the prevalence of psychological distress in all participants according to sex, age (≤64 vs ≥65 years), jobless status (unemployed vs other), and degree of social contact (often/sometimes vs not often/rarely) are shown in Figure 2. A statistically significant decreasing trend in prevalence was noted between 2005 and 2009 for all groups except those reporting infrequent social contact (P = 0.0659). In relation to the respective comparison groups, the prevalence of psychological distress was higher among those reporting infrequent social contact ( Figure 2, lower right) and unemployed participants (Figure 2, lower left) and tended to be higher among women than among men ( Figure 2, upper left). There was no significant difference in prevalence between age groups (Figure 2, upper right).

RESULTS
Despite the overall decreasing trend in prevalence, psychological distress appeared to increase from 2008 to 2009 (Figure 1), although the change (+3.7%) was not statistically significant (P = 0.1590). However, changes in prevalence within some subgroups ( Figure 2) were statistically significant, such as the increases in the prevalence of psychological distress among women (+7.5%; P = 0.0403) and participants aged 65 years or older (+7.2%; P = 0.0400).

DISCUSSION
This study presented findings related to recovery from psychological distress among adults living in a community hit by a large earthquake. We found that the prevalence of psychological distress decreased gradually and reached its lowest point at 4 years after the earthquake. Although the short-term effects of earthquakes on mental health are well-documented, there are limited data on the long-term impacts of such events. Most prior studies assessed psychological disorders several months or a few years after earthquakes, but few studies investigated longterm outcomes (ie, >5 years after an earthquake). 12, 13 Bland et al 12 assessed symptoms of psychological distress in people 7 years after the 1980 earthquake and found that levels of psychological distress were higher among those who had reported damage from the earthquake than among those who had not. Priebe et al 13 assessed mental disorders in a randomly selected sample of 200 people who had experienced a 1997 earthquake in Italy 8 years before and concluded that the prevalence of mental disorders did not differ from that in comparable populations. Although the results of these 2 studies are inconsistent and not directly comparable with ours, the findings suggest that a large earthquake could affect the mental health of victims for a very long time, up to 7 years. Thus, to facilitate comparison with previous studies, further follow-up investigation in Yamakoshi is warranted.
The present subgroup analysis showed that the prevalence of psychological distress did not decrease over 5 years among those with poor social contact. This finding agrees with the results of a number of previous reports on earthquakes, which showed that disruption of social networks is a major factor in poor mental health. Oyama et al 11 found that earthquake victims who were estranged from others in their community were at higher risk (OR = 7) for psychological distress 3 years after the Niigata-Chuetsu earthquake. Bland et al 14 reported that those who were evacuated farther away from family or friends had higher levels of psychological distress 3 to 4 years after an earthquake in Italy. Kiliç et al 15 showed that disruption of social networks due to relocation increased psychological distress 4 years after 2 severe earthquakes in Turkey.
The prevalence of psychological distress was higher among unemployed people, which was also noted in other studies. 11,16 This may be due in part to financial burdens from house damage that adversely affect psychological health. 17 The prevalence of psychological distress was higher among women than among men in 2005, 2006, and 2009. In general, common mental disorders are more prevalent among women, 18 but a sex difference in the prevalence of GHQdefined psychological distress was not found in a large Japanese population. 19 A possible explanation for this higher prevalence in women may be the traditional, dependent roles of women in rural Japanese communities, including Yamakoshi. 11,20 In such communities, women have a major role at home: they manage most of the housework and care for all family members, including their husband, children, and husband's parents. 11 Because of role change, post-earthquake burdens may have been greater for women than for men. The long-term higher prevalence of psychological distress in women versus men is noteworthy in relation to suicide rate after earthquakes. Long-term suicide rates after the Niigata-Chuetsu earthquake increased among women and decreased among men. 21 The suicide rate among women may have been elevated because the prevalence of psychological distress in women was higher and increased over time.
The prevalence of psychological distress generally decreased after the earthquake, although there was an unexpected increase from 2008 to 2009. While multiple logistic regression analysis did not identify independent factors that contributed to this increase, there might have been stress-related factors beginning in 2009. Many people had been struggling to reestablish their lives and had difficulty continuing their daily tasks, which may have been reflected in greater psychological distress in 2009.
Many earthquake victims may have had difficulty rebuilding their lives due to financial losses, which are common after a large earthquake. 12,22 In fact, the Niigata prefectural government reported declines in per capita income in the disaster areas after the Niigata-Chuetsu earthquake. 23 Per capita income in the disaster areas decreased in 2008-2009, after rising in 2005-2007 as a result of temporary financial aid from the government (Figure 3). 23 Financial problems caused by the earthquake may have been an important contributor to the worsening mental health of some Yamakoshi people 5 years later.
We evaluated the prevalence of psychological distress in an earthquake-stricken area rather than in a normal population. Few studies have used the GHQ-12 to investigate the prevalence of psychological distress in normal Japanese populations. Using the same GHQ-12 cutoff score, of 3/4, Shimbo et al 24 reported that the prevalence rate of psychological distress was 18.9% to 20.0% in a randomly selected general population (n = 600) aged 20 to 69 years. This suggests that the prevalence of psychological distress in the present population was higher than that in a normal Japanese population.
A major limitation of this study is that participation rates differed throughout the 5-year study period; those between 2006 and 2008 (around 50%) were lower than those in 2005 and 2009. This discrepancy may have influenced prevalence rates. Secondly, we did not assess psychological distress before the earthquake and thus cannot determine whether the prevalence of psychological distress returned to preearthquake levels. Finally, the residents of Yamakoshi are mostly elderly adults in a rural area and, hence, generalization of our results requires caution. Our findings might not be applicable to younger populations in urban areas. Despite these limitations, data on long-term changes in postearthquake psychological status among people living in a community near an earthquake epicenter are rare and worth reporting.
In summary, the overall prevalence of psychological distress steadily decreased during the 4-year period after the earthquake; however, prevalence tended to increase between the fourth and fifth years after the earthquake. In addition, the prevalence of psychological distress did not decrease among those who reported poor social contact. Rebuilding of the lives of earthquake victim is ongoing; thus, their mental health status should be closely monitored.

ONLINE ONLY MATERIALS
Abstract in Japanese.