INTRODUCTION
Suicide is a global public health problem, being the leading cause of death worldwide, accounting for approximately 800,000 deaths annually [1]. In an effort to address this issue, researchers and clinicians have developed several scales to assess factors associated with high suicide risk for suicide prevention. However, generalizing the characteristics of suicide attempters remains challenging when relying solely on certain existing measures [2].
The Risk-Rescue Rating Scale (RRRS) was developed as a clinical scale to measure the severity of self-harm at the time of a suicide attempt, and the lethality of a suicide attempt could be calculated as a function of the physical risk and rescue of the attempt [3]. Higher levels of risk combined with lower levels of rescue increase the fatality rate of suicide attempts, and characterization of each factor is an important variable in assessing suicide attempt risk [4]. Factor analysis of physical risk may be able to estimate the type and prognosis of suicide attempts, and factor analysis of rescue may provide useful guidance for developing guidelines on how to increase the rescue of suicide attempters.
Many countries have announced policies to prevent the spread of coronavirus disease-2019 (COVID-19), from blocking travel between countries, to social distancing, to locking down entire cities [5,6]. These policies have significantly affected people’s lifestyle in many ways, which has also become a factor influencing mental health and suicide attempts [7,8]. A number of previous studies have reported neuropsychological changes, including anxiety, depression, and cognitive impairment, and increased rates of depressive and anxiety disorders due to restrictions on social activities during the pandemic [9,10]. Further, several mental health experts have warned of an increase in actual suicides and a pandemic-like spread of suicide attempts during large-scale pandemics like COVID-19 [11,12]. Increased or new risk factors for suicide, such as social isolation, loneliness, infection, fear and uncertainty of information overload, economic problems, and domestic violence, have been cited by people during pandemics as reasons for the increase in suicides, and previous large-scale pandemics of severe acute respiratory syndrome and Spanish flu have supported this argument [12-14].
While there has been much research on the link between the COVID-19 pandemic and suicide, only few studies have analyzed the physical risk factors of suicide attempters or analyzed factors related to rescue. There is also a lack of research on how suicide is affected by the unique environment of social and physical isolation attributed to the spread of the pandemic during COVID-19.
In summary, the objectives of the current study are twofold. First, we used the RRRS to characterize the degree of quarantine intensification during COVID-19 and the physical risk and rescue factors of suicide attempters attending emergency medical centers. Second, we aimed to determine the effect of social isolation through social distancing policy in South Korea during the COVID-19 pandemic on suicide attempts to identify the changed characteristics of suicide attempts.
METHODS
Participants and procedure
This study examined patients who attempted suicide and visited a regional emergency medical center, Soonchunhyang University Bucheon Hospital. The research method involved reviewing and analyzing the medical records of suicide attempters who visited the center during the study period, as well as the consultation records of case managers from the Life Love Crisis Response Team. Of the 2,215 people who visited the emergency department (ED) during the study period, 937 participants were excluded due to difficulties in accurately interviewing them, those who simply had suicidal thoughts, and those with missing data, resulting in a total of 1,278 suicide attempts. Participants were then categorized according to the duration of social distancing, with 392 before, 543 during, and 343 after (Figure 1).
We excluded psychiatric emergency patients who visited the regional emergency medical center of Soonchunhyang University Bucheon Hospital from January 1, 2019 to December 31, 2022, if they were missing data corresponding to the RRRS to be analyzed, had only suicidal thoughts and not suicide attempts, or could not be accurately interviewed. The study sample was categorized into three groups based on the duration of social distancing: before, during, and after it ended (Figure 1). The Institutional Review Board approved the study (IRB No. 2023-03-014-001).
Categorizing study
Social distancing period
To determine the parameters of the study period, we defined the following three different social distancing periods according to the government guidelines and policies. First, we defined before social distancing period from January 1, 2019 to February 28, 2020 given that the official launch of the “social distancing” campaign took place on March 1, 2020. Second, we defined during social distancing period from March 1, 2020 to December 31, 2021. Third, we defined the end of the social distancing period as January 1, 2022 to December 31, 2022 as social distancing given that since November 2021, social distancing has been fluidly implemented, tightened, and loosened depending on the situation, making it difficult to define a clear timeframe.
Measures
RRRS
The RRRS is a validated quantitative tool used to evaluate the severity of suicide attempts. Consisting of 10 items, each scored from 1 to 3, it meticulously assesses both risk and rescue elements inherent in such events. Risk factors include aspects such as method used, degree of impairment of consciousness, level of toxicity, reversibility, and treatment required (Table 1). Conversely, rescue factors examine the location, initiator of the rescue, likelihood of detection, accessibility of help, and time to detection (Table 2). By calculating the difference between the rescue and risk scores, the resulting Risk-Rescue Score effectively quantifies the lethality of the attempt (Figure 2).
The RRRS uses a comprehensive scoring system to differentiate between risk and rescue factors, providing insight into the potential lethality of suicide attempts. Each item is carefully scored on a scale of 1 to 3, with higher scores indicating increased risk or greater likelihood of rescue. Importantly, while a lower rescue score indicates an increased likelihood of lethality, risk factor scoring works in the opposite direction. To streamline the analysis, the initial risk and rescue scores are converted to a 5-point scale, allowing for a more nuanced understanding of the lethality of suicide attempts. Ultimately, the scale provides a precise framework for evaluating the interplay between risk and rescue elements, enabling clinicians and researchers to make informed assessments of suicide attempt lethality.
Statistical analysis
Continuous variables in the study are presented as mean and standard deviation, while nominal variables are presented as frequencies (percentages, %). The study used SPSS version 22 for Windows (IBM Corp.) for data analysis, with a significance level of p<0.05 and a confidence interval (CI) of 95%. The characteristics of individuals presenting to the ED during the study period, including the COVID-19 social isolation phase, were examined, including demographic factors such as age, sex, marital status, and cohabitation status, as well as aspects related to the suicide attempt, including method difficulty and time to detection. In addition, the relationship between social distancing and suicide attempts categorized by physical risk according to the RRRS was examined using binary logistic regression to derive odds ratios (ORs). We also analyzed the association between social distancing and physical risk of suicide attempts, examining each risk factor of the RRRS separately. For detailed analysis, a crosstabulation method was used to compare baselines before, during, and after the implementation of social distancing interventions. In addition, the relationship between social distancing and suicide attempts classified by rescue potential according to the RRRS was examined using binary logistic regression to calculate odds ratios. Finally, the study determined the underlying cause of the association by cross-analyzing each rescue factor of the RRRS independently.
RESULTS
General characteristics of suicide attempters
Table 3 shows the general characteristics of individuals who attempted suicide and presented to the ED. The mean age of the three groups, categorized by social distancing policy, was compared and found to be 37.4 (±18.09) years old in the social distancing group, which was lower than the mean age of 40.9 (±18.18) years old before the implementation (p=0.018).
The study included 693 (54.2%) male participants and 585 (45.8%) female participants. After the implementation of social distancing, 111 (32.4%) male participants and 232 (67.6%) female participants were recorded, indicating a higher proportion of females (p<0.001). Regarding marital status, single individuals had the highest proportion of 646 (50.5%) overall. The association between social distancing policies and marital status was statistically significant. The proportion of single individuals increased from 173 (44.1%) before social distancing to 286 (52.7%) during social distancing and to 187 (54.5%) after social distancing (p=0.002). The proportion of individuals who had a cohabitant throughout the entire period was higher (68.4%) compared to those who did not. The association between cohabitation status and coronavirus distancing policies was statistically significant (p<0.001). After social distancing was implemented, the proportion of noncohabitants increased significantly to 42.0% (144 individuals) during routine distancing, which was a significant increase from the previous two groups. The risk of suicide attempt method, which is used to assess physical risk, had the highest rate of 1,008 (78.9%) in the low-risk group. The association with corona distancing policy showed an increasing pattern of suicide attempt method in the low-risk group: 277 (70.7%) before the implementation of the three periods, 441 (81.2%) during social distancing, and 290 (84.5%) after the implementation. Meanwhile, the rate decreased in the other moderate and severe risk groups (p<0.001). The study analyzed the time of discovery after a suicide attempt and its association with the distancing policy. The highest proportion of 505 (39.5%) was discovered between 1 and 4 hours. The results showed that social distancing was associated with early discovery decreasing to 122 (22.5%), moderate discovery to 207 (38.1%), and late discovery to 214 (39.4%) compared to before the implementation (p=0.001).
Associations between social distancing and suicide attempts by risk during the ‘RRRS’
Figure 3 shows the OR results for the association between social distancing and the risk of any of the RRRS factors for suicide attempts. There was a statistically significant increase in low-risk suicide attempts during social distancing (OR, 1.67; 95% CI, 1.13-2.46) and an increase in low-risk suicide attempts during everyday distancing (OR, 2.39; 95% CI, 1.47-3.90) based on the period before the social distancing policy was implemented. There was a statistically significant decrease in high-risk suicide attempts during social distancing (OR, 0.60; 95% CI, 0.41-0.89) and after implementation (OR, 0.41; 95% CI, 0.26-0.68).
Analysis of changes in the risk of suicide attempts following the implementation of social distancing
To examine the association between social distancing policies and the risk of suicide attempts, we analyzed the risk factors of the RRRS before social distancing was implemented (baseline) and during social distancing, and the results are shown in Table 4. This provides a detailed analysis of the factors shown in Figure 3.
During the social distancing period, the level of risk increased to 488 cases (89.9%) of low risk and 55 cases (10.1%) of high risk, compared to before the implementation (p=0.012). Additionally, moderate suicide attempts, such as drowning, suffocation, and hanging, decreased to 91 cases (16.8%) during the implementation period, compared to the pre-implementation period (p<0.001). Regarding toxicity and lesions, the number of severe high-risk factors decreased by almost half to 90 (16.6%) during the implementation period compared to 118 (30.1%) in the period before implementation (p<0.001). For reversibility factors, complete recovery increased to 368 (67.8%) during implementation from 183 (46.7%) before implementation, and the time required for complete recovery decreased to 149 (27.4%) from before implementation (p<0.001). Finally, during implementation, the level of care significantly improved with 393 (72.4%) cases requiring only emergency room treatment compared to 184 (46.9%) before implementation. Additionally, there was a significant reduction of more than half in the number of cases requiring intensive care, with only 57 (10.5%) cases during implementation (p<0.001).
To investigate the association between social distancing policies and the risk of suicide attempts, we analyzed the risk factors of the RRRS before and after the implementation of social distancing. The results are presented in Table 5. The factors of the periods before and after the introduction of social distancing are analyzed in detail (Table 5) in order to confirm the factor analysis of the results in Figure 3.
After implementation, the level of risk decreased to 25 (7.3%), which is half the pre-implementation rate (p<0.001). The number of moderate suicide attempts by drowning, suffocation, and hanging decreased significantly from 107 (27.3%) before implementation to 47 (13.7%) after implementation (p<0.001). There was no significant difference in the level of consciousness in Table 4, However, Table 5 indicates a difference, with a decrease in the number of high-risk coma cases from 24 (6.1%) to 11 (3.2%) after implementation, and an increase in the number of lucid cases to 253 (73.8%) with a higher rate (p=0.019). Toxicity and lesions more than doubled pre-implementation, with 147 (42.9%) mild cases and 40 (11.7%) severe cases, halving the pre-implementation rate of 118 (30.1%) (p<0.001). Table 4 shows a significant difference in reversibility, but Table 5 shows no statistically significant difference. Finally, regarding treatment, the percentage of cases that only required minor first aid increased significantly to 263 (76.7%). In contrast, the percentage of high-risk cases that required intensive care decreased to 28 (8.2%), which is a one-third decrease from 98 (25.0%) before implementation (p<0.001).
Changes in the rescue of suicide attempters by social distancing (via RRRS screening)
Figure 4 displays the OR results for the association between social distancing and the rescue component of the RRRS for suicide attempts. There was an increase in low-rescue suicide attempts during social distancing (OR, 1.44; 95% CI, 1.11-1.87) based on the period before the social distancing policy was implemented. After the policy was implemented, there was a statistically significant decrease in low-rescue suicide attempts (OR, 0.38; 95% CI, 0.27-0.52). There were statistically significant decreases in high-rescue suicide attempts during social distancing (OR, 0.70; 95% CI, 0.54-0.90) and increases in high-rescue suicide attempts during routine distancing (OR, 2.65; 95% CI, 1.91-3.68).
Examination of changes in the rescue of suicide attempts following social distancing
The aim of this study was to investigate the association between social distancing policies and suicide attempters’ rescue factors. The rescue factors of the RRRS were analyzed before social distancing was implemented (baseline) and during social distancing. The results are presented in Table 6, which confirms the factor analysis of the results in Figure 4. The factors of rescue difficulty, discoverability, and rescueability were found to be statistically significant. During the social distancing period, there were 273 (50.3%) low-rescue suicide attempts compared to 162 (41.3%) before implementation. This resulted in a larger proportion of low-rescue attempts than high-rescue attempts (p=0.007). Additionally, discoverability showed that 136 (25.0%) took steps to avoid accidental discovery during social distancing, which is more than double the pre-implementation rate of 50 (12.8%) (p<0.001). Finally, regarding rescueability, 93 out of the total suicide attempts during social distancing (17.1%) requested help, which is almost half the pre-implementation rate of 125 (31.9%). Meanwhile, 294 suicide attempts (54.1%) did not seek help, and 156 (28.7%) left only clues, resulting in lower scores (p<0.001).
To investigate the relationship between the social distancing policy and the structural factors of suicide attempters, the structural factors of the RRRS were analyzed. Table 7 is a factor analysis of the results in Figure 4, before and after implementation.
At the level of rescue, the proportion of low-rescue suicide attempts was halved after implementation to 72 (21.0%) compared with 162 (41.3%) before implementation (p<0.001). This shows that the number of high-rescue suicide attempts increased, in contrast to (Table 6). In terms of location, there was a significant decrease in the proportion of moderately unfamiliar locations, such as a car or home, from 45 (11.5%) before implementation to 20 (5.8%) after implementation, and a shift towards suicide attempts in familiar locations to 315 (91.8%) (p=0.023). Rescuers were more likely to be close family and acquaintances than before implementation, with 248 (72.3%) self-reported cases in Table 6 and 10 (1.8%) self-reported cases in Table 7, but the proportion of self-reported cases more than doubled to 26 (7.6%) in Table 7. Bystander detection halved from 21 (5.4%) before implementation to 7 (2.0%) after implementation (p=0.001). In terms of detectability, cases detected due to concealment or carelessness compared to before implementation (baseline), contrary to the results in Table 6, decreased 12-fold in Table 7, from 50 cases (12.8%) before implementation to 3 cases (0.9%) after implementation. The number of cases of trying to avoid detection even in familiar spaces was halved from 120 (30.6%) before implementation to 49 (14.3%) after implementation (p<0.001). The opposite result was found for the likelihood of rescue (Table 6). The number of suicide attempters calling for help increased from pre-implementation to 121 (35.3%), and the number of suicide attempters leaving only clues decreased from 70 (17.9%) pre-implementation to 34 (9.9%) post-implementation (p=0.009). Finally, detection time actually decreased after implementation, with 95 (27.7%) rapid detections compared to 153 (39.0%) pre-implementation, and a shift towards moderate detections (p=0.001).
DISCUSSION
This study examined the impact of social distancing on suicide attempters by analyzing changes in physical risk and rescue characteristics. The risk and rescue characteristics of suicide attempters were assessed using a specialized measure, the RRRS. We found that social distancing policies were significantly associated with suicide attempters’ rescue. We also found changes in risk.
In analyzing the effect of social distancing on general characteristics of suicide attempters in this study, we found that age decreased from a mean age of 40.9 (±18.18) before social distancing to a mean age of 37.4 (±18.09) during social distancing during the COVID-19 pandemic. This supports previous research showing that suicide attempts among younger age groups (12-19 year sand 20-29 years) increased to the highest levels during the COVID-19 pandemic compared to suicide attempts among other age groups, suggesting that social isolation affected younger age groups who enjoyed active social lives [15], contributing to the increase in suicide attempts among younger generations. Previous research suggests that younger suicide attempts are more impulsive than older suicide attempts and that younger people are less serious about suicide, preferring low-risk suicide attempts with less medical lethality [16], which may explain the increase in low-risk suicide attempts in this study.
In terms of sex, males attempted suicide at a higher rate than females throughout the study period, with an even greater difference during the social distancing period. While most previous studies have shown an increase in suicide attempts among women and adolescents and younger age groups [17-19], this study found an increase in suicide attempts among men. This may be due to economic reasons, such as difficulties in finding employment due to the COVID-19 pandemic, unemployment or regional characteristics, but it is also possible that the increase in suicide attempts among men is due to the regional characteristics of the study area, which may not be representative of women’s suicide attempts.
A final discussion of the general characteristics of suicide attempters is the cohabitation factor. Across all time periods, a higher proportion of suicide attempts occurred in cohabiting households than in single-person households. Contrary to previous research suggesting that cohabitation is one of the protective factors against suicide attempts, we found a higher proportion of suicide attempts in cohabiting households [20]. This suggests that even with a cohabiting partner, there may be isolation within the household that can lead to feelings of disconnection. This suggests that the higher rate of suicide attempts in cohabiting households than in single-person households may be due to increased time spent together as a result of enforced closure, which may lead to increased discord, including conflict, argument and violence, which may contribute to increased suicide attempts [21].
This study found that there were 1.67 times more low-risk suicide attempts during implementation than before implementation and 2.39 times more low-risk suicide attempts after implementation than before implementation (baseline), indicating a gradual increase in low-risk suicide attempts compared to before implementation of social distancing.
When the factors contributing to this increase in low-risk suicide attempts were analyzed using the risk factors of the RRRS, there were significant differences in method, toxicity, reversibility and outcome of suicide attempts during the social distancing period compared with the pre-implementation period.
In the post-implementation period, the factors that contributed to the increase in low-risk suicide attempts were method of attempt, toxicity and treatment outcome, in addition to level of consciousness was added.
Most previous studies have focused on the physical risk of patients during the policy transition, with few studies identifying social and cultural risk factors, such as perceived isolation, chronic illness, unemployment and family discord, that increase the risk of suicide attempts during an infectious disease pandemic [22-26].
Other studies identifying risk factors for suicide attempts have focused on identifying risk factors for high-risk suicide attempts by scoring risk and structural factors using the RRRS and calculating mortality rates, or on identifying psychological factors [2,27-29]. This is the opposite of the current approach to suicide attempts, which focuses only on the high-risk group and is therefore far from finding prevention methods in the current situation where low-risk suicide attempts are increasing. Therefore, although high-risk suicide attempts are of course important and should be prevented, it is important to propose new solutions for the prevention of low-risk suicide attempts, as the number of low-risk suicide attempts has increased in the last four years. This is not limited to low-risk suicide attempts; first-time suicide attempts or mild suicide attempts are increasingly likely to lead to repeated suicide attempts, often of higher intensity, and eventually to high-risk suicide attempts [30,31].
It is therefore necessary to analyze the risk factors of low-risk suicide attempts in order to modify prevention policies.
In this study, the change in rescue suicide attempts in response to changes in COVID-19 social distancing periods was significantly different for each intervention. During social distancing, there was a 1.44-fold increase in low rescue suicide attempts from pre-implementation (baseline) and a 2.65- fold increase in high rescue suicide attempts from post-implementation (baseline). This suggests that social isolation is an important factor influencing suicide attempts, with lower levels of rescue during the social distancing period and higher levels of rescue in the post-implementation period. Previous studies have also found an increase in low rescue [24], but this study was able to identify the specific factors that contribute to low rescue.
When analysing the factors of increased low rescue during the social distancing period in terms of RRRS rescue, we found significant differences in the factors of detectability and rescueability. The likelihood of detectability seemed to double as the characteristics of the suicide attempters changed from easily detectable suicide attempts in front of witnesses to more accidental or hidden suicide attempts due to reluctance to be found. In terms of rescueability, we found that the number of suicide attempters who sought help with high rescueability characteristics decreased by about half during the social distancing period, shifting to characteristics that made rescue more difficult.
The increase in low rescueability during the social distancing period may also be influenced by the fact that suicide attempters transported by ambulance may be uniquely affected by COVID-19, as previous studies have shown that emergency medical services bottlenecks and hospital care limitations, including delays in dispatch time due to processes to prevent infection during ambulance transport, delays in transport due to the explosion in demand for transport of patients with respiratory infections with limited resources, and increased waiting times on arrival at the hospital [32]. Therefore, the impact of COVID-19 on suicide attempters presenting to the ED by ambulance requires further study.
After implementation of social distancing, we found higher levels of resilience than before implementation (baseline), and analyzes of the ‘RRRS’ resilience subfactors showed significant differences in all factors: location, rescuer, discoverability, rescueability and time of discovery.
Social isolation, such as during a pandemic, can occur at any time, suggesting that during these periods, suicide attempts should be monitored for forms of reduced rescue, such as accidental or hidden attempts to avoid detection, or significantly reduced calls for help.
The limitations of this study are that it is a retrospective, single-centre analysis with limited generalisability and subject to recall bias. However, the study was conducted in a metropolitan area of the Republic of Korea, which has the size and population distribution of a large city, and has sufficient reliability as a study with diverse population characteristics. In addition, we did not identify socio-cultural factors such as psychological or other economic reasons that may influence suicide attempts, but this study aimed to identify the physical and rescue factors of suicide attempts within the important limitation of social distancing during COVID-19 and did not attempt to analyze other general causes. Under the factor of rescue, we have summarised the fragmented content of the 119 ambulance service during the social distancing period of COVID-19, but we have not analyzed its impact on suicide attempts in detail. Therefore, future research on the impact of COVID-19 on the 119 ambulance service is needed. A final limitation is that the RRRS used in this study is an outdated scale, first created in 1972, and its format has not changed significantly. However, it was used because it is the only formalised measure of risk of physical harm and rescue in suicide attempts. There is therefore a need for further development and refinement of the scale to include more recent trends.
Nevertheless, this study was the first attempt to analyze the changes in the morphology of suicide attempts in the special situation of COVID-19 and to analyze in detail the characteristics of risk and rescue of suicide attempts, and it is hoped that identifying the changed factors of suicide attempts in large-scale social isolation such as infectious diseases like COVID-19 will help to develop programmes for the prevention of suicide attempts.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
