The COVID-19 pandemic has necessitated the implementation of preventive measures, which have had a detrimental impact on mental health (1–3). Additionally, psychiatric services have faced significant reductions (4). Persons with severe mental illnesses (SMI) have shown to be highly sensitive to the changes brought by the COVID-19 pandemic (5).

Unfortunately, Croatia experienced two major earthquakes during the pandemic. On March 22nd, 2020, the capital of Croatia, Zagreb was struck by an earthquake measuring 5.5 on the Richter scale, causing extensive damage to numerous healthcare facilities (6). On December 29th, another earthquake measuring 6.4 on the Richter scale occurred near Zagreb, specifically in Petrinja, which is approximately 40 kilometers away (7). The impact of earthquake can exacerbate addictive behavior and influence the factors related to mental ill-health and the utilization of mental health services (8–10). Double disasters, such as the pandemic and earthquakes in this case, pose a unique challenge to the mental health of populations (11, 12). Globally, changing conditions directly affect the complexity of current and future disaster management issues (11). Previous research has underscored the necessity for special attention and long-term support in addressing the psychological impacts of double disasters (12, 13). Higher incidence of posttraumatic stress disorder, depression, and anxiety have been reported following such disasters (12, 13). Certain pre-existing psychiatric conditions, such as obsessive-compulsive disorder, may exhibit significant clinical worsening (14). Experiencing loss, displacement from one’s place of residence, socio-economic challenges, lack of social support, and negative coping mechanisms, such as substance abuse, can induce psychological stress (13). Moreover, persons with SMI are particularly prone to experiencing higher levels of stress compared to the general population (15). They also inherently express a higher rate of certain psychoactive substances and behavioral addictive disorders (16). Discontinuation of pharmacological therapy may further exacerbate their condition (13). Therefore, the occurrence of a double disaster can potentially place persons with mental illnesses in an extremely vulnerable position.

The primary aim of this research is to examine changes in substance use and addictive behavior in persons with SMI during the first and second double disasters—the COVID-19 waves and the co-occurring earthquakes. Furthermore, we seek to explore mental ill-health, coping mechanisms, psychiatric treatment, and perceived social support as predictors of changes in substance use and addictive behavior.

The sampling is shown in Figure 1. From overall 169 RECOVER-E participants, 90 participants engaged in the first measurement of this study. The assessment included 47 participants included in the CMHT group and 43 in TAU group. Altogether six participants dropped out in the second measurement: two in the CHMT group and 4 in the TAU group. There was no statistically significant difference between the control and intervention group (TAU vs. CMHT group) in socio-demographic and medical data, except in the use of long acting injectables (LAI) (CMHT group received LAIs more than TAU group). We refer the reader to another publication with more details about the population (35).

Table 1 shows baseline socio-demographic and medical data of the population and scores of questionnaires exploring the presence of substance or behavior addiction disorder, mental ill-health, coping mechanisms, and perceived social support in the first and second study point.

Within the sample, approximately 2.4% of participants (N = 2) were identified with harmful alcohol consumption and 3.4% (N = 3) with alcohol dependence using AUDIT, 1.2% (N = 1) with probable drug abuse using DUDIT and 5.8% (N = 5) with probable gambling problem using SOGS in the first assessment. In the second assessment, approximately 2.4% (N = 2) were showing harmful use of alcohol and 1.2% (N = 1) of participants were identified as alcohol dependent, 1.2% (N = 1) were having probable drug abuse, while nobody identified as a pathological gambler. Since increased alcohol consumption, psychoactive substance use, gaming, and gambling measured by AUDIT, DUDIT and SOGS were reported by less than 5 persons, we did not include it in the further analyses.

The major increase was observed in the use of tobacco smoking and sedatives (Table 2). All data about self-reported changes in substance use and addictive behavior are presented in Table 2.

A higher DASS-21 score was a positive predictor of increased tobacco use in the second assessment (p = 0.021, OR 1.102, 95% CI 1.015–1.197). In the first assessment, increased sedative use was predicted in participants not receiving CMTH service, i.e., those in TAU group (p = 0.031, OR 8.225, 95% CI 1.208–56.015), by discontinuation in antipsychotic medications (p = 0.010, OR 0.020, 95% CI 0.001–0.386) and by higher PSS score (p = 0.011, OR 1.216, 95% CI 1.046–1.413). Higher ISI and MSPSS family subscale scores predicted higher sedative use in the second assessment (p = 0.007, OR 1.222, 95% CI 1.057–1.413; p = 0.034, OR 0.576, 95% 0.346–0.959, respectively). All results remained statistically significant even after adjustment with the FDR set at 5%. Details are shown in Table 3.

The results of our study revealed a noteworthy increase in tobacco smoking and the use of sedatives among individuals with SMI during both instances of the double disaster. It is important to note that the frequency of psychoactive substance use in this sample was notably lower than what has been reported in previous literature (36). This disparity can be attributed to several factors, including the implementation of anti-COVID-19 measures, such as the “stay-at-home” national policy, which disrupted conventional sources of drugs and led to an observed increase in pricing (37). Drug availability is one of the important risk factors for addiction (38). Additionally, specific cultural standards regarding addiction within the socio-cultural model may have biased self-reporting (39).

Despite the relatively small number of individuals exhibiting harmful use and behavior in our sample, a significant increase in smoking and sedative use was evident at both study points, mirroring trends in the general population (40). Notably, a higher prevalence of tobacco smoking was already documented among individuals with SMI in the pre-pandemic time (41), amplifying their health burden (42). The increase in sedative use is also a cause for concern, given its association with a higher risk of mortality (43).

The use of tobacco was predicted by depression and anxiety symptoms and discontinuation of antidepressant use, whereas increased use of sedatives was predicted by insomnia, not- receiving CMHT treatment, discontinuation of antipsychotics, and perceived levels of stress and social support. High environmentally – induced stress contributes to mental-ill health and elicits different, sometimes maladaptive coping responses, including the use of substances (44). It is worth noting that environmental stressors, which can contribute to mental ill-health and maladaptive coping responses, including substance use, played a significant role in the context of the COVID-19 pandemic and earthquakes (44, 45). The documented increase in depression, anxiety, and stress symptoms during these challenging times aligns with findings from other studies (46–48). Importantly, depression and anxiety symptoms have known associations with addictive behavior (46), while stress is independently associated with SMI (49). This heightened stress level not only exacerbates the risk of substance use and addictive behavior but also underscores the importance of effective mental health interventions, particularly in times of crisis.

Our study found that participants who reported receiving greater social support were at a lower risk of engaging in substance use. This finding aligns with previous research that has shown a significant relationship between the lack of social support and an increased risk of addiction during the pandemic (36). Social support plays a crucial role in promoting recovery (50) and can act as a protective factor against the negative impact of stigma and shame. This, in turn, has a positive effect on an individual’s quality of life and mental health (51).

Psychiatric treatment has been shown to have a significant effect on tobacco and sedative use. Participants receiving CMHT treatment had a lower risk for substance use. CMHTs as an outreach service have been recognized as a valuable resource for persons with SMI and addictive disorders for many years (52). The availability of mental health services, such as CMHTs, has been shown to have multiple benefits. It can enhance treatment adherence, alleviate anxiety associated with the loss of service support, and provide essential medical advice (53). Notably, in some situations, CMHTs were the only accessible psychiatric service for individuals with SMI during the pandemic and earthquakes in Croatia (10). This highlights their pivotal role in ensuring continued care for the most severely affected patients, presenting a novel approach to healthcare during crises (54, 55). Furthermore, it is recommended that CMHTs should not be limited to pandemic times but should be sustained and expanded beyond such emergencies (56, 57). This model of care should be developed and implemented in countries where it does not currently exist (56, 57). There is a scarcity of studies examining the impact of the organization of psychiatric services, including CMHT services, on individuals with SMI during a pandemic or natural catastrophe. While a few examples exist (11, 12, 35, 58), further research in this area is warranted to better understand their role and effectiveness in crisis situations.

The discontinuation of antidepressant medications emerged as a predictor of increased tobacco use in our study. This finding is noteworthy considering that effective treatments for smoking cessation are currently available (59), and efforts to raise awareness about cessation strategies during the pandemic have been initiated (60). These results may indicate a potential gap in healthcare. Similarly, the increase in sedative use was predicted by the discontinuation of antipsychotic treatment. This could be explained by the shortfall of sleep induction (61). The discontinuation of psychiatric treatment is attributable to the lock-down measures and the disruption of the standard care in the assessment (35).

These results highlight the profound impact of external stressors on the unhealthy lifestyles of individuals with SMI. As mentioned earlier, individuals with SMI are inherently vulnerable, and additional burdens can significantly compromise their well-being (42, 43). Unfortunately, the anticipated rise in the frequency of natural disasters in the coming years (62) necessitates proactive measures.

During and after a natural disaster, the mental and physical health of marginalized populations, including those with SMI, is particularly at risk (63). Clinicians must be well-prepared and equipped with the skills and knowledge required to deliver continuous psychiatric care to individuals with SMI, especially in cases of double disasters. This includes the capacity to provide mental health services in the community, tailored to the specific needs of the SMI population, and in coordination with other essential public services, such as social services, labor services, and housing. Additionally, policymakers must prioritize emergency preparedness and response strategies tailored to the needs of vulnerable populations during crises (63). Given the far-reaching impacts of natural disasters, cross-national responses may be necessary. This was especially evident during the pandemic, underlining the crucial role played by international associations and informal organizations in alleviating the effects of traumatic events. They have the potential to develop scientifically universal guidelines and algorithms specifically designed for persons with SMI, applicable to various domestic circumstances at every level. With well-structured preventive measures in place for persons with SMI, the risks to their well-being can be significantly mitigated.

In conclusion, in a case of a double disaster, additional health burden due to tobacco smoking and sedative use in persons with SMI needs to be foreseen. To counteract the effect, special attention needs to be given to reducing mental-ill health and stress, providing social support and continuity of psychiatric care, through both medications and CMHTs.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by the Ethics Committee of University Hospital Centre Zagreb. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

SM: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing, Writing – original draft. IRP: Supervision, Writing – review & editing. JG: Investigation, Writing – original draft, Project administration, Writing – review & editing. SL: Conceptualization, Data curation, Investigation, Methodology, Project administration, Writing – review & editing. LS-Z: Funding acquisition, Resources, Supervision, Writing – review & editing. FB: Software, Writing – review & editing. ZB: Investigation, Writing – review & editing. ZM: Investigation, Project administration, Writing – review & editing. IF: Writing - review and editing, Supervision. MR: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.