On April 26, 1984, Unit 4 of the Chernobyl Nuclear Power Plant suffered several explosions and fires, which released large amounts of radionuclides into the surrounding environment (1). Nearby populations were evacuated two days later, yet, the radioactive particles spread far beyond the initial evacuation zone, eventually contaminating an area with a population of greater than five million (1, 2). To this day, the Chernobyl nuclear disaster is considered one of the largest and most disastrous of all nuclear accidents (2, 3).

Populations affected by large-scale disasters are often prone to increased rates of mental health disorders, such as posttraumatic stress and depression, and are more likely to self-medicate through alcohol or substance abuse (4–6). Additionally, disaster-surviving populations self-report worse physical and mental health than similar non-disaster counterparts (7). This pattern is not surprising; a poor self-rating of personal health often correlates with poor psychosocial health (8). These changes in mental and physical health are theorized to be caused not only by the immediate stressor of a disaster itself but also by stress due to sudden evacuation, unexpected danger, unknown negative health effects, changes or difficulties with living situations, and familial conflict (9, 10).

What is especially unique about nuclear disasters is the ambiguity and uncertainty of the strength and extent of danger. Unlike other natural disasters like hurricanes or wildfires, where signs of damage are clear and visual or tactile, nuclear disasters have no obvious indicators of threat or danger.

Populations living nearby disaster areas lack answers to whether they are exposed and, if so, how severely. We suspect this ambiguity is critical in the development of mental health disorders in the aftermath of nuclear disasters.

With nuclear disasters, victims also bear the additional burden of stigmatization as radioactively contaminated persons (9, 11). As media coverage labels those exposed as victims, they become seen as contaminated or dirty (11). This stigma or ostracism can also lead to increased levels of social stress and can be yet another potential contributor to a higher risk of mental health problems (12).

Dramatic or significant stressors like disasters may be related to the precipitation or relapse of various mental health disorders according to the diathesis–stress model. The diathesis–stress model assumes that all individuals demonstrate various likelihoods for developing mental health disorders, whether due to genetics or environmental pressures. Under enough stress, a threshold of tolerance and one’s ability to cope can be surpassed and a mental health disorder develops (13, 14). We hypothesize that the Chernobyl nuclear disaster and ensuing stressors are adequately traumatic to be associated with increased rates of mental health disorders, especially since man-made technological disasters, specifically nuclear ones, are the most terrifying of all disasters (15, 16).

Higher rates of poor mental-health are common in populations exposed to nuclear radioactivity. Survivors of the nuclear weapons unleashed on Hiroshima and Nagasaki had higher rates of anxious and somatic symptoms decades after the bombings, especially for those living closer ground zero (17, 18).

In the Gomel region, an area of Belarus near the Chernobyl disaster, investigators used the General Health Questionnaire to measure local psychopathological symptoms. A sample from the region had much higher rates of anxious and depressive symptoms than a demographically similar group living at a greater distance from Chernobyl. However, the difference in prevalence rates of clinically diagnosable psychiatric disorders between the two groups was not statistically significant (19).

Clean-up workers (liquidators) exposed to radiation during the remediation process, or at minimum the threat of it, experienced higher rates of diagnosable mood and anxiety disorders compared to controls, yet, did not struggle significantly more with alcohol disorders nor intermittent explosive disorder [IED (20)]. Among Estonians, even 24 years after the incident, those involved in cleanup measures were three times more likely to suffer unexplained symptoms of somatization and were more likely to struggle with alcohol abuse than controls. Sleep problems were also significantly elevated as well as symptoms of agoraphobia (21).

A study involving Russian immigrants to the United States found that those who lived within 150 km of the Chernobyl disaster prior to emigration had much higher scores on the Russian Beck Anxiety Inventory and the Russian Beck Depressive Inventory than those who lived greater than 150 km from Chernobyl (22). Finally, a concluding review of several studies confirmed that populations in Belarus and Ukraine that were exposed to Chernobyl experienced greater incidences of depression, anxiety, and medically unexplained symptoms, but at levels which were not clinically diagnosable (9).

Chernobyl exposure was associated not only with poor mental health of surrounding populations, but perception of personal health was also worse. In the Russian immigrant study, those who lived closer to the site of the accident self-reported poor mental health at three times the rate of those further from the disaster (22). Another study in the Gomel region described that those who lived closer to Chernobyl self-reported poor health at much greater rates than an unaffected, distant Russian community, even though on a clinical scale no definite distinctions between the overall health of groups were observed (23).

Many of the studies undertaken thus far have been scientifically rigorous, but several shortcomings in the literature need to be addressed. Most projects have examined the mental health effects of the Chernobyl disaster effectively, but with convenience and non-representative samples; our study complements many of their objectives, but with a large, population-based representative sample of the entirety of Ukraine. Second, our survey tool allows us to make clinically based psychiatric diagnosis for various disorders based on DSM-IV qualifications. Finally, our study enables the scientific community to draw conclusions about the long-term effects of the disaster due to the 16-year gap between the accident (1986) and survey implementation (2002). Information about the long-term effects of nuclear disasters on mental health can provide insight for care workers and government agencies in future or current disaster remediation processes, like those working with the survivors of the 2011 Fukushima Daiichi nuclear disaster.

With information from the Ukraine World Mental Health Survey (Ukraine-WMH), we examined the effect of self-reporting that one had lived “in the zone contaminated as a result of the Chernobyl accident” (24, 25) on prevalence and relapse rates of psychological disorders and self-reported respondent health scores. Comparisons between groups were examined by using binomial logistic regression modeling and chi-square tests. Given highly differential rates of mental health disorders between sexes in Ukraine (26), as wells as the possibility of gender acting as a diathesis for different disorders, we looked at overall models and models stratified by sex.

Our analyses showed that some rates of clinically diagnosable DSM-IV mental health disorders are more prevalent in Ukrainians who reported that they lived in the zone affected by the Chernobyl disaster than Ukrainians who did not report that they were affected by the disaster. Specifically, prevalence of alcohol use disorders and IED was higher for those who lived in the zone affected by Chernobyl when compared to non-affected controls. By gender, men were at a greater risk of having alcohol disorders, while women experienced an increased risk of IED. Finally, individuals who reported living in the Chernobyl-affected zone rated their overall physical and mental health significantly worse than those who lived elsewhere in Ukraine. Under the theoretical framework of the diathesis–stress model, we conclude that the threat of exposure to radionuclides, whether real or imagined, may have been adequately stressful for those with prior diatheses to develop psychiatric disorders, although due to the nature of disaster research and our dataset, we are unable to make causal statements about this robust association.

Our results agree with the findings of previous studies, which conclude that rates of mental health disorders are likely to rise in populations affected by disaster (3–6, 8), and specific studies that report higher levels of distress and mental anguish from the Chernobyl disaster and others like it (7, 9–11, 17–23, 28, 29). Our findings regarding increased rates of alcohol use disorders and IED, and our lack of stimulating findings for anxiety disorders are interesting, especially considering previous research, which suggests anxiety disorders may be the most susceptible class of disorders to large-scale traumatic events (4, 9, 21–23, 29–33). Perhaps one reason anxiety disorders were not identified at a significantly higher rate among those reporting Chernobyl exposure compared to controls is due to our failure to include PTSD in our analyses. Unfortunately, PTSD was not assessed on all participants who took the Ukraine WHM-CIDI; the resulting small sample size hindered its inclusion in the anxiety disorders analysis.

Furthermore, we found that self-reported health ratings were significantly lower for the population exposed to Chernobyl. This finding confirms previous conclusions about self-reported health in other of disaster-surviving populations (8), as well as other studies focused specifically on the Chernobyl disaster (7, 22, 23).

Although inclusive for a wide range of health-related concerns, our study focused only on psychiatric disorders and mental health, rather than physical outcomes from radionucleotide exposure. The physical effects of the disaster and subsequent radioactive contamination have been well-documented elsewhere (1, 34).

One interesting aspect of our investigation was the time elapsed from disaster to data collection. Data collection took place in 2002, 16 years after the disaster. With such a large window between the disaster and data collection, the presence of disorders due to Chernobyl or the memory of them might have faded. Yet, this elapsed time was also an advantage, giving us the opportunity to observe the long-term effects of the disaster beyond several months or years.

A limitation of our analyses is that the false negative (respondents who said they did not live in the contaminated zone but did) and false positive (respondents who reported living in the contaminated zone but did not) rates are unknown. The survey instrument also did not inquire about severity of exposure. Furthermore, our cross-sectional design limits our understanding of the process or timeframe through which subjects developed mental health disorders or coped. Even with these limitations, however, we found significant differences in rates of mental health disorders associated with perceived exposure to the Chernobyl disaster.

Our findings confirm the conclusions of other related studies, but add scientific rigor to the conversation surrounding the effects of Chernobyl by using a population-based representative sample of Ukraine, a survey instrument with WHO-approved DSM-IV diagnostic capabilities, and a long-term timescale.

Given ours and others’ results concerning the Chernobyl disaster (1, 7, 9, 19–23) and research on the effects of nuclear weapons on public mental health (17, 18), current patterns that researchers have observed recently in Fukushima survivors (35) might continue for another decade, if not more. Our results may prove helpful in predicting the long-term outcomes of other nuclear disasters. If so, we, like Kunii et al. (35), encourage ongoing concern and monitoring for those affected by the Fukushima nuclear disaster. Further discussion and understanding of the ambiguous threats of nuclear disasters on public mental health can educate care workers and government agencies in preparation for future nuclear disasters, and guide continued care for those still suffering from these tragedies.

This study was carried out in accordance with the recommendations of the Committees on Research Involving Human Subjects of Stony Brook University as well as the Kiev International Institute of Sociology (KIIS) and the Ukrainian Psychiatric Association (UPA) internal review boards, with written informed consent from all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the Committees on Research Involving Human Subjects of Stony Brook University as well as the KIIS and the UPA internal review boards.

NT, LH, and MB developed the research plan, and MB performed statistical analysis and drafted initial versions of the manuscript. All authors approved the final version of the manuscript.

This research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.