Suicide is a highly complex phenomenon and the fourth leading cause of death among people aged 15–29 years ( 1,2,3). Because of relatively high rates of mental health issues, young adults (18–25 years) seem to be at particular risk (4, 5). Research indicates that as many as 10.9 % of young adults have experienced a major depressive episode, as opposed to 7.4 % reported in adults aged 26–49 years and 4.8 % in those aged 50 years or older (6, 7). In addition, the age-specific prevalence of depression is the highest in those aged 18 to 24 years (21.5 %) and the lowest in those aged ≥65 years (14.2 %) (8). Among people who have committed or attempted suicide the most common mental disorder is some kind of mood disorder (25 % and 21 %, respectively), followed by schizophrenia (8 % and 7 %, respectively) (9). The most common method of suicide attempt among young adults is self-poisoning, most often with readily available legal prescription drugs such as analgesics, benzodiazepines, atypical antipsychotics, and antidepressants (10).
With an average suicide rate of about 19 per 100,000 inhabitants, Serbia is one of the leading European countries in this respect (11), especially since the 13.5 % increase in mental health issues coinciding with the United Nations economic sanctions, NATO bombing, refugee and displacement issues, and social unrests between 1999 and 2002 (12). The year 2020 seems to have worsened the issue with the COVID-19 pandemic, as 10.8 % of students reported feeling depressed, while symptoms of depression, anxiety, and stress among university students increased significantly at the peak of the pandemic (9.9 %, 26.4 %, and 16.2 %, respectively) (13, 14). At the same time, emergency departments reported a 60 % (spring 2020), 10 % (summer 2020), and 30 % (winter 2021) increase in interventions over suicide attempts compared to 2019 (15). Risk factors for attempted suicide during COVID-19 included female gender, younger age, and depressive disorder (16, 17).
Given the vulnerability of young adults, the aim of our retrospective cohort study was to look deeper into the demographic and clinical profile and the characteristics of committed and attempted suicides by poisoning in young adult Serbian population during the COVID-19 pandemic.
Our study included data about 193 patients aged 18–25 years, hospitalised over intentional self-poisoning at the Clinic for Emergency and Clinical Toxicology of the National Poison Control Centre (NPCC), Military Medical Academy (MMA) in Belgrade, Serbia between 2020 and 2023, as this four-year period coincides with the COVID-19 pandemic.
We analysed their anonymised electronic medical records entered into the NPCC database, which include data collected from admission to discharge. Beside the age of 18–25 years, the inclusion criteria were hospitalisation over intentional self-poisoning (X60–X69) according to the International Statistical Classification of Diseases and Related Health Problems (ICD-10) (18). Excluded were hospitalised patients outside the age range, outpatients of any age, and patients received over accidental (unintentional) poisoning.
The study was approved by the Ethics Committee of the MMA (Decision No. 65/2024 ) and conducted in accordance with the Declaration of Helsinki and its amendments (19).
Data were analysed using SPSS v. 22.0 statistical software (IBM Corp., Armonk, NY, USA).
Demographic, clinical, and self-poisoning data are presented with absolute and relative frequencies, arithmetic means and standard deviations (SD). To determine the normality of distribution we used the Kolmogorov-Smirnov or the Shapiro-Wilk test. Relationships between qualitative variables were determined with the chi-squared test or Fisher’s exact test when the expected frequencies were <5 or for 2×2 contingency tables. Significant chi-squared results for crosstabs larger than 2×2 were further examined with post-hoc pairwise comparisons of column proportions using the Bonferroni correction. The Mann-Whitney U test was used for continuous variables. The correlations between the patients’ clinical profile and characteristics of suicide attempt was assessed with Pearson’s correlation coefficient. Statistical significance was set at p<0.05.
Table 1 shows the demographic and clinical profile of 193 young adult emergency ward patients who attempted suicide by poisoning and characteristics of their suicide attempts. Even though 53 had no specified mental disorder diagnosis on record, we included them as having a mental disorder, because the record showed previous psychiatric treatment. Self-harm recorded for 15 patients refers to self-inflicted injuries on body surface, either by cutting (in 13 patients) or by hitting and banging their head (in two patients). The terms ”organic diseases” and “organic medications” refer to diseases other than psychiatric and prescription drugs other than those used in psychopharmacotherapy. Complications associated with self-poisoning include pneumonia and other respiratory tract diseases, urinary infections, central nervous system disorders, cardiovascular complications, and gastrointestinal symptoms.
Demographic and clinical profile of young adult emergency ward patients who attempted suicide by poisoning and characteristics of suicide attempts (N=193)
| N (%) or mean ± SD | ||
|---|---|---|
| Gender | female | 107 (55.4) |
| male | 86 (44.6) | |
| Age (years) | 21.60±2.20 | |
| Year of self-poisoning | 2020 | 39 (20.3) |
| 2021 | 49 (25.4) | |
| 2022 | 49 (25.4) | |
| 2023 | 56 (28.9) | |
| Current mental disorders | 109 (56.5) | |
| Categories of current mental disorder | depressive disorder | 22 (11.4) |
| neurotic disorder | 7 (3.6) | |
| psychotic disorder | 5 (2.6) | |
| personality disorder | 8 (4.1) | |
| addiction | 14 (7.3) | |
| NA | 53 (27.5) | |
| Current psychopharmacotherapy | 79 (40.9) | |
| Categories of current psychopharmacotherapy | benzodiazepines | 45 (23.3 ) |
| antidepressants | 41 (21.2) | |
| antipsychotics | 49 (25.4) | |
| mood stabilisers | 46 (23.8) | |
| Self-harm | 15 (7.8) | |
| Organic diseases | 34 (16.1) | |
| Organic medications | 21 (10.9) | |
| Characteristics of suicide attempts by poisoning | ||
| Duration of hospitalisation (days) | 2.94±3.77 | |
| Number of attempts | 1 | 130 (67.4) |
| ≥2 | 46 (23.8) | |
| NA | 17 (8.8) | |
| Substances used | Prescription drugs | 137 (71.0) |
| PAS (illegal drugs / alcohol) | 4 (2.1) | |
| Prescription drugs + PAS | 33 (17.1) | |
| Other (corrosives, insecticides, acid) | 9 (4.7) | |
| Prescription drugs + other (corrosives, insecticides, acid) | 10 (5.1) | |
| Number of substances used | 1 | 68 (35.2) |
| ≥2 | 125 (64.8) | |
| Prescription drugs used in self-poisoning | benzodiazepines | 109 (56.5) |
| antidepressants | 28 (14.5) | |
| antipsychotics | 50 (25.9) | |
| affective stabilizers | 67 (34.7) | |
| cardiovascular | 29 (15.0) | |
| analgesics | 38 (19.7) | |
| History of previous attempts at self-poisoning with prescription drugs | 38 (19.7) | |
| Reason for self-poisoning | personal issues | 68 (35.2) |
| family issues | 18 (9.3) | |
| health issues | 2 (1.0) | |
| NA | 104 (53.9) | |
| Complications of self-poisoning | 66 (34.2) | |
| Outcome of self-poisoning | non-fatal | 188 (97.4) |
| fatal | 5 (2.6) |
NA – data not available; PAS – psychoactive substances; SD – standard deviation
Of the five patients who died, three were male and two female. The youngest among them was 18 and the oldest 25 years old. All had attempted suicide for the first time and used not only legal, prescription drugs but corrosive substances and illicit drugs as well. Their toxicological-chemical blood analysis revealed the presence of valproate [630 mg/L (therapeutic concentrations 50–100 mg/L)] in the first deceased patient; concentrated sodium acids in the second; enalapril [10.82 mg/L (therapeutic concentrations 0.01–0.1 mg/L)], bromazepam [0.71 mg/L (therapeutic concentrations (0.05–0.2 mg/L)], and nifedipine [0.38 mg/L (therapeutic concentrations 0.01–0.2 mg/L)] in the third; 3,4-methylenedioxymethamphetamine [1.34 mg/L (coma and fatal concentrations 1–4.2 mg/L)] in the fourth; and venlafaxine [1.25 mg/L (therapeutic concentrations (0.06–0.4 mg/L)], klonazepam [0.3 mg/L (therapeutic concentrations (0.004–0.07 mg/L)], and olanzapine [1.2 mg/L (therapeutic concentrations (0.001–0.08 mg/L)] in the fifth.
There is no significant difference in the number of young adults who attempted suicide by poisoning between the study years (chi-squared=3.041, df=3, p=0.385), but using two or more substances was significantly more common than using one substance alone (chi-squared=86.622, df=2, p=0.000).
Table 2 shows differences between the genders. Briefly, female young adults were significantly younger than male and had a history of poisoning attempts more often, yet had fewer complications than male patients.
Gender distribution in relation to the demographic and clinical profile and characteristics of suicide attempts by poisoning in young adult patients (N=193)
| Demographic and clinical profile | N (%) or mean ± SD | Z / chi-squared test | p-value | ||
|---|---|---|---|---|---|
| Male (n=86) | Female (n=107) | ||||
| Age | 22.16±2.12 | 21.15±2.16 | −3.143 | 0.002 | |
| Current mental disorders | no | 32 (37.1) | 33 (30.8) | 0.058 | 0.810 |
| yes | 45 (52.2) | 64 (59.8) | 1.082 | 0.298 | |
| NA | 9 (10.7) | 10 (9.4) | 2.372 | 0.124 | |
| Current psychopharmacotherapy | no | 35 (40.7) | 40 (37.3) | 0.221 | 0.638 |
| yes | 34 (39.5) | 45 (42.1) | 0.122 | 0.726 | |
| NA | 17 (19.8) | 22 (20.6) | 0.020 | 0.889 | |
| Self-harm | no | 80 (93.0) | 97 (91.6) | 0.372 | 0.542 |
| yes | 6 (7.0) | 9 (7.5) | 0.152 | 0.697 | |
| NA | 0 (0.0) | 1 (0.9) | 0.810 | 0.368 | |
| Characteristics of suicide attempts by self-poisoning | |||||
| Number of self-poisoning attempts | 1 | 64 (74.4) | 66 (61.7) | 3.534 | 0.060 |
| ≥2 | 13 (15.1) | 33 (30.8) | 6.502 | 0.011 | |
| NA | 9 (10.5) | 8 (7.5) | 0.533 | 0.465 | |
| Year of self-poisoning | 2020 | 19 (22.1) | 20 (18.7) | 0.336 | 0.562 |
| 2021 | 18 (20.9) | 31 (29.0) | 1.638 | 0.201 | |
| 2022 | 20 (23.3) | 29 (27.1) | 0.372 | 0.542 | |
| 2023 | 29 (33.7) | 27 (25.2) | 1.664 | 0.197 | |
| Duration of hospitalisation (days) | 3.63±5.20 | 2.39±1.88 | −1.266 | 0.206 | |
| Reason for self- poisoning | personal | 28 (32.6) | 40 (37.4) | 0.490 | 0.484 |
| family | 7 (8.1) | 11 (10.3) | 0.260 | 0.610 | |
| health | 1 (1.2) | 1 (0.9) | 0.026 | 0.873 | |
| NA | 50 (58.1) | 55 (51.4) | 1.124 | 0.289 | |
| Number of substances used | 1 | 26 (30.2) | 42 (39.3) | 1.700 | 0.226a |
| ≥2 | 60 (69.8) | 65 (60.7) | |||
| History of previous self-poisoning attempts with prescription drugs | no | 65 (75.6) | 71 (66.4) | 1.960 | 0.162 |
| yes | 11 (12.8) | 27 (25.2) | 4.666 | 0.031 | |
| NA | 10 (11.6) | 9 (8.4) | 0.563 | 0.453 | |
| Complications of self-poisoning | no | 49 (57.0) | 78 (72.9) | 5.370 | 0.023a |
| yes | 37 (43.0) | 29 (27.1) | |||
| Outcome of self-poisoning | fatal | 3 (3.5) | 2 (1.9) | 0.495 | 0.657a |
| non-fatal | 83 (96.5) | 105 (98.1) | |||
SD – standard deviation;
Fisher’s exact test
Table 3 shows a significant positive correlation of current mental disorder and current psychopharmacotherapy with the overall number of suicide attempts by poisoning per patient, history of previous self-poisoning, and the number of substances used. There is also a positive correlation between self-harm and the number of suicide attempts and substances used.
Pearson’s correlation between the clinical profile and characteristics of suicide attempts by poisoning in young adult patients (N=193)
| Characteristics of suicide attempts by self-poisoning | Clinical profile | ||||
|---|---|---|---|---|---|
| Current mental disorders | Current psychopharmacotherapy | Self-harm | Organic diseases | Organic medications | |
| Number of suicide attempts | 0.283** | 0.317** | 0.176* | 0.083 | 0.020 |
| History of previous suicide attempts with prescription drugs | 0.269** | 0.268** | −0.071 | 0.095 | 0.053 |
| Number of substances used | 0.304** | 0.288** | 0.166* | 0.086 | 0.089 |
| Complications of self-poisoning | 0.188 | 0.106 | 0.053 | −0.086 | −0.049 |
Significant Pearson’s correlations are in boldface:
p<0.05;
p<0.01
The Pearson correlation coefficient also revealed a positive correlation between self-harm and current mental disorders (r=0.200, p=0.008).
To the best of our knowledge, this is the first study in Serbia to specifically analyse young adults who attempted suicide by poisoning. Our findings show that more than half suffered from a mental disorder and that they mostly used prescription drugs (benzodiazepines in particular) to commit suicide. This is in line with earlier research in adult Serbian population identifying benzodiazepines as the most common drug in suicide attempts (20). In fact, the rate of benzodiazepine prescriptions is very high in Serbia (21). This is also in line with reports claiming that adolescents under the age of 18 attempt suicide with paracetamol and ibuprofen (22, 23), where as older adults resort to prescribed drugs with worse outcomes (24). It is also important to note that more than half of our patients used a combination of two substances or more.
More young women than men had had a history of previous self-poisoning attempts, while young men suffered from more complications. These findings are consistent with the “gender paradox” in suicide, i.e., that women attempt suicide more often than men but use less lethal means or doses, which suggests that adverse psychosocial factors play a greater role in suicide in men (25,26,27). Our findings are also similar to earlier reports for populations of a similar age (26, 28, 29, 30).
Among mental disorders, depressive disorders prevailed in both genders. However, these findings are limited, as no diagnosis data were available for nearly a third of the patients registered as having some kind of mental disorder. Before the COVID-19 pandemic, a significant increase was reported in serious psychological distress among young adults in the US between 2008 and 2018, with nearly half experiencing and nearly 40 % receiving treatment for mental health issues (31, 32,33). However, since the pandemic broke out, there was a significant increase in intentional self-poisonings among 13–15-year-old girls but not among young adults (22, 34).
We had expected a higher prevalence of self-poisoning with illegal drugs (psychoactive substances) and alcohol (2 %), given the 2018 reports of alcohol and illicit drug use among over a third of young adult Americans (35). In England, Tyrrell et al. (36) found that 7.6 % of patients aged 10–24 years used opioids and 4.2 % used psychostimulants for self-poisoning (36). Instead, we found a higher prevalence (17.0 %) of those attempting suicide by combining alcohol, psychostimulants, and prescription drugs.
Furthermore, current mental disorder and current psychopharmacotherapy significantly correlate with the number of suicide attempts by poisoning, previous history of suicide attempts with prescription drugs, and the number of substances involved in self-poisoning. The same is true for the correlation between self-harm and the number of suicide attempts by poisoning. Similar correlations have been reported by Lumpe et al. (37), especially between current mental disorders (mainly depressive disorder) and repeated suicide attempts, but their age span is quite larger than ours. They also found that most patients used their own legally prescribed drugs, usually in combination.
Our study is retrospective, i.e., data were collected from available medical records, and certain information is lacking, most notably on specific mental disorders in 53 patients. Another limitation is that the period before the COVID-19 pandemic is not part our study, which makes a comparison and some insight into possible trends impossible. To overcome these limitations, further research should be prospective and follow up surviving patients through every stage of recovery, from initial treatment on hospital admission to home care.
Our results suggest that the experience of the pandemic has highlighted the need to strengthen mental healthcare in Serbia. This could be achieved by developing resilience in young people, establishing help centres, and encouraging populations at risk to seek professional help in case of a crisis through national information campaigns.
To be effective, suicide prevention programmes should include raising public awareness of the issue through both traditional media and social networks. Psychiatric assessment and support should become mandatory components of treatment. Additionally, psychosocial interventions and long-term monitoring can help lower the risk of repeated self-poisoning.