Trends in suicide deaths before and after the COVID-19 outbreak in Korea.

We investigated the effect of the coronavirus disease-2019 (COVID-19) pandemic on suicide trends in Korea via a time-series analysis. We used Facebook Prophet to generate forecasting models based on the monthly numbers of suicide deaths in Korea between 1997 and 2018, validated the models by comparison with the 2019 numbers, and predicted the numbers of suicides in 2020. We compared the expected and observed numbers of suicides during the COVID-19 pandemic. The total numbers of suicides during the COVID-19 pandemic did not deviate from projections based on the pre-pandemic period. However, the number of suicides among women and those under the age of 34 years significantly exceeded the expected level. The COVID-19 pandemic did not increase the overall suicide rate significantly. However, suicides among women and young people increased, suggesting that the pandemic might drive more members of these groups to suicide. Further studies are needed to verify the long-term impact of the COVID-19 pandemic on suicide.

1. Introduction

Since early 2020, the coronavirus disease 2019 (COVID-19) pandemic has been a public health crisis and has overwhelmed people’s daily lives globally [1]. People are fearful of the high infectivity and mortality of COVID-19 and restrict their daily activities, such as meeting family and friends, exercising, and visiting hospitals [2]. To curtail the spread of COVID-19, authorities have implemented stringent social distancing policies, including restricting public gatherings, closing schools, and requiring the wearing of face masks in public facilities [3]. The prolonged pandemic has aggravated financial difficulties for low-income earners due to increased unemployment and reduced income [4]. As a result, people continue to experience stress while adapting to the extreme circumstances of the pandemic [5, 6].

Given this situation, experts are concerned about the deterioration of people’s mental health and the consequent increase in suicide rate [7]. Psychological distress due to isolation, loneliness, bereavement, increased alcohol consumption, financial stressors, and disrupted health care are major factors that could increase the suicide risk during the COVID-19 pandemic [8]. However, recent research could not find evidence of an increase in suicide rates during the pandemic [9, 10]. A recent study of global changes in suicide trends before and after the COVID-19 outbreak showed that suicide numbers remained largely unchanged or even declined in high- and upper-middle-income countries [11]. This interrupted time-series analysis found statistical evidence of a declining trend in suicides in the early months of the pandemic in Korea. However, this finding was calculated from preliminary data for a short period in the early months of the pandemic, and care should be taken in its interpretation.

Although the prevalence of COVID-19 in Korea has been lower than other countries and the Korean government did not institute a complete lockdown, the social distancing policies and restricted economic activities have persisted [12]. Furthermore, the stress, anxiety, and depression levels of Koreans during the pandemic were greatly increased compared with before the pandemic, as in other countries [13, 14]. Before the COVID-19 outbreak, suicide was one of the most serious public health and social issues in Korea, where the suicide rate is the highest among Organization for Economic Cooperation and Development countries [15, 16]. Therefore, suicide trends during the COVID-19 pandemic in Korea need to be closely monitored. Accordingly, this study investigated changes in suicide incidence before and after the COVID-19 outbreak in Korea in more detail, examining the difference between the number of suicides during the pandemic (February to December 2020) and the projections based on data obtained over the past two decades.

2. Materials and methods

2.1. Data sources and study population

Using the cause of death statistics provided by Statistics Korea’s Microdata Integrated Service, we obtained data on the sex, age, and date of death of 283,633 suicide victims between 1997 and 2020 [17]. The codes assigned to the cause of death due to suicide were X60–X84 according to the International Statistical Classification of Diseases and Related Health Problems 10th Revision. From these data, we calculated the number of suicides per month by sex and age group (≤ 34, 35–49, 50–64, and ≥ 65 years).

This study was approved by the Institutional Review Board of Chonnam National University Hospital (IRB number: CNUH-EXP-2021-047).

2.2. Time series analysis

From data on suicide deaths each month between 1997 and 2019, 264 observations up to December 2018 were used to fit the time series models forecasting the next 24 months. The 12 observations for 2019 were used to validate the models.

We used Facebook Prophet, a time series forecasting procedure based on an additive model in which non-linear trends are fitted with seasonality and holiday effects [18]. Prophet is as sophisticated as the autoregressive integrated moving average (ARIMA) model, a widely used forecasting procedure [19]. Prophet has the advantage of being able to adjust for the effects of holidays and other recurring events, which is why we used the Prophet algorithm to adjust for the effect of celebrity suicides around the COVID-19 outbreak. First, using the Prophet algorithm, we fitted a time series model based on data for 264 months between 1997 and 2018 with multiplicative seasonality mode. The effect of 19 celebrity suicides was modelled with extra regressors. Then, we tuned the hyperparameter to match the observed and expected numbers in 2019 as closely as possible in terms of the root mean squared error, mean absolute error, and mean absolute percentage error, and made predictions with a 24-month forecasting period (the expected number of suicides per month between 2019 and 2020). Additionally, when comparing accuracy for predicting suicides in 2019 between the Prophet and ARIMA algorithms, we determined that the fitted Prophet models performed comparably to, or better than, the ARIMA models (S1 Table). Finally, we visualized the observed and expected number of suicides.

We compared this expected number to the observed number of suicides during the COVID-19 pandemic (February to December 2020) by calculating observed-to-expected ratios (OERs) and confidence intervals (CIs). Statistical significance was defined as a 95% CI that excluded the null value of 1.00. All analyses were conducted using R (ver. 4.0.3; R Development Core Team, Vienna, Austria).

3. Results

3.1. Observed and expected numbers of suicides per month during the COVID-19 pandemic

Figs 1–3 depict the number of suicide deaths per month between 2017 and 2020. Tables 1–3 show the observed and expected numbers of suicides per month during the pandemic, together with OERs and 95% CIs. The total number of suicides per month during the COVID-19 pandemic did not deviate overall from the projected estimates (Fig 1 and Table 1). The monthly number of male suicides was slightly lower than expected (Fig 2(A)), but the difference was not significant (Table 2). However, the monthly number of female suicides greatly exceeded the expected level (Fig 2(B) and Table 2). Compared to the projections, the actual number of suicides per month during the pandemic was lower among those aged 50–64 years (Fig 3(C) and Table 3), but significantly higher among those aged ≤ 34 years (Fig 3(A) and Table 3). The monthly number of suicides by those aged 35–49 (Fig 3(B) and Table 3) and ≥ 65 (Fig 3(D) and Table 3) years was largely in line with the projections.

Fig 1

Observed and expected numbers of suicides per month between 2017 and 2020 in Korea.

The red line with circles represents the observed number of suicides between 2017 and 2020. The blue line with triangles represents the expected number of suicides according to forecasting models and the surrounding blue area indicates the 95% confidence interval (with 2019 as the validation period). The blue vertical dotted lines indicate the dates of celebrity suicides.

Fig 3

Observed and expected numbers of suicides by age group per month between 2017 and 2020 in Korea.

The red line with circles represents the observed number of suicides between 2017 and 2020. The blue line with triangles represents the expected number of suicides according to forecasting models and the surrounding blue area indicates the 95% confidence interval (with 2019 as the validation period). The blue vertical dotted lines indicate the dates of celebrity suicides.

Table 1

Number of suicides per month during the COVID-19 pandemic in Korea.

	Observed	Expected (95% CI)	OER (95% CI)
Feb	993	1,033.67 (822.03–1,253.58)	0.96 (0.90–1.02)
Mar	1,149	1,098.95 (875.38–1,303.30)	1.05 (0.99–1.11)
Apr	1,120	1,111.68 (910.28–1,322.49)	1.01 (0.95–1.07)
May	1,152	1,162.50 (942.11–1,379.77)	0.99 (0.93–1.05)
Jun	1,135	1,067.91 (850.25–1,281.66)	1.06 (1.00–1.12)
Jul	1,228	1,220.71 (1012.15–1,443.11)	1.01 (0.95–1.06)
Aug	1,185	1,158.28 (944.42–1,373.32)	1.02 (0.96–1.08)
Sep	1,020	1,038.64 (818.93–1,258.88)	0.98 (0.92–1.04)
Oct	1,147	1,016.27 (808.98–1,227.47)	1.13 (1.06–1.19)
Nov	1,061	1,143.76 (922.67–1,357.29)	0.93 (0.87–0.98)
Dec	913	958.46 (740.14–1,155.45)	0.95 (0.89–1.01)

Bold indicates a statistically significant OER with a 95% CI that excludes the null value of 1.00.

Abbreviations: OER, Observed-to-Expected Ratio; CI, Confidence Interval.

Table 3

Number of suicides by age group per month during the COVID-19 pandemic in Korea.

	Age ≤ 34			Age 35–49			Age 50–64			Age ≥ 65
	Observed	Expected (95% CI)	OER (95% CI)	Observed	Expected (95% CI)	OER (95% CI)	Observed	Expected (95% CI)	OER (95% CI)	Observed	Expected (95% CI)	OER (95% CI)
Feb	181	172.14 (99.73–248.33)	1.05 (0.90–1.20)	303	289.80 (219.87–366.97)	1.05 (0.93–1.16)	272	316.80 (257.66–372.72)	0.86 (0.76–0.96)	237	255.23 (198.77–305.56)	0.93 (0.81–1.05)
Mar	222	171.11 (105.09–243.24)	1.30 (1.13–1.47)	304	301.52 (228.42–376.54)	1.01 (0.89–1.12)	371	347.30 (293.01–403.07)	1.07 (0.96–1.18)	252	283.93 (225.55–341.57)	0.89 (0.78–1.00)
Apr	218	165.34 (95.65–232.11)	1.32 (1.14–1.49)	293	287.87 (213.93–362.82)	1.02 (0.90–1.13)	325	358.22 (303.14–418.89)	0.91 (0.81–1.01)	284	309.47 (251.48–367.19)	0.92 (0.81–1.02)
May	234	170.51 (101.35–242.26)	1.37 (1.20–1.55)	294	307.27 (236.94–384.96)	0.96 (0.85–1.07)	349	363.94 (311.01–422.05)	0.96 (0.86–1.06)	275	325.47 (271.85–384.42)	0.84 (0.75–0.94)
Jun	237	158.91 (90.94–233.16)	1.49 (1.30–1.68)	271	277.42 (197.45–349.29)	0.98 (0.86–1.09)	314	327.66 (270.00–358.24)	0.96 (0.85–1.06)	313	309.47 (250.46–367.39)	1.01 (0.90–1.12)
Jul	253	191.27 (124.03–264.49)	1.32 (1.16–1.49)	326	328.83 (255.82–403.36)	0.99 (0.88–1.10)	333	381.03 (320.80–439.07)	0.87 (0.78–0.97)	316	332.84 (275.08–390.19)	0.95 (0.84–1.05)
Aug	211	169.78 (103.41–235.10)	1.24 (1.08–1.41)	295	326.84 (255.61–400.31)	0.90 (0.80–1.01)	332	370.65 (312.02–424.35)	0.90 (0.80–0.99)	347	305.97 (252.25–361.93)	1.13 (1.01–1.25)
Sep	210	150.78 (83.74–218.61)	1.39 (1.20–1.58)	254	277.47 (200.91–355.99)	0.92 (0.80–1.03)	272	343.59 (287.20–400.41)	0.79 (0.70–0.89)	284	282.35 (226.67–340.70)	1.01 (0.89–1.12)
Oct	218	148.00 (76.29–216.79)	1.47 (1.28–1.67)	276	269.42 (196.29–350.41)	1.02 (0.90–1.15)	327	319.79 (264.02–377.21)	1.02 (0.91–1.13)	326	287.58 (231.51–342.28)	1.13 (1.01–1.26)
Nov	238	211.55 (144.73–279.37)	1.13 (0.98–1.27)	255	325.91 (250.47–398.07)	0.78 (0.69–0.88)	285	330.73 (275.49–386.75)	0.86 (0.76–0.96)	283	261.78 (203.19–314.49)	1.08 (0.96–1.21)
Dec	168	170.12 (102.19–241.82)	0.99 (0.84–1.14)	270	263.24 (187.53–343.28)	1.03 (0.90–1.15)	258	299.53 (243.78–353.85)	0.86 (0.76–0.97)	217	221.55 (164.34–275.65)	0.98 (0.85–1.11)

Bold indicates a statistically significant OER with a 95% CI that excludes the null value of 1.00.

Abbreviations: OER, Observed-to-Expected Ratio; CI, Confidence Interval.

Fig 2

Observed and expected numbers of male and female suicides per month between 2017 and 2020 in Korea.

The red line with circles represents the observed number of suicides between 2017 and 2020. The blue line with triangles represents the expected number of suicides according to forecasting models and the surrounding blue area indicates the 95% confidence interval (with 2019 as the validation period). The blue vertical dotted lines indicate the dates of celebrity suicides.

Table 2

Number of male and female suicides per month during the COVID-19 pandemic in Korea.

	Men			Women
	Observed	Expected (95% CI)	OER (95% CI)	Observed	Expected (95% CI)	OER (95% CI)
Feb	714	753.84 (602.94–907.40)	0.95 (0.88–1.02)	279	289.99 (197.48–377.51)	0.96 (0.85–1.07)
Mar	783	796.02 (639.15–944.73)	0.98 (0.91–1.05)	366	314.10 (230.87–402.58)	1.17 (1.05–1.28)
Apr	764	812.96 (664.94–956.05)	0.94 (0.87–1.01)	356	310.28 (217.86–400.68)	1.15 (1.03–1.27)
May	791	847.27 (703.44–998.12)	0.93 (0.87–1.00)	361	327.66 (237.80–411.82)	1.10 (0.99–1.22)
Jun	761	774.45 (628.74–915.65)	0.98 (0.91–1.05)	374	309.65 (220.11–396.87)	1.21 (1.09–1.33)
Jul	849	881.98 (752.26–1,024.15)	0.96 (0.90–1.03)	379	320.80 (238.81–410.32)	1.18 (1.06–1.30)
Aug	810	833.53 (686.18–977.81)	0.97 (0.90–1.04)	375	315.40 (229.32–401.42)	1.19 (1.07–1.31)
Sep	685	770.31 (625.85–925.09)	0.89 (0.82–0.96)	335	276.24 (191.23–356.64)	1.21 (1.08–1.34)
Oct	795	728.10 (587.92–880.08)	1.09 (1.02–1.17)	352	299.22 (207.24–386.25)	1.18 (1.05–1.30)
Nov	715	774.86 (624.23–924.99)	0.92 (0.86–0.99)	346	372.46 (277.15–460.37)	0.93 (0.83–1.03)
Dec	655	680.92 (527.01–838.26)	0.96 (0.89–1.04)	258	282.94 (195.56–369.58)	0.91 (0.80–1.02)

Bold indicates a statistically significant OER with a 95% CI that excludes the null value of 1.00.

Abbreviations: OER, Observed-to-Expected Ratio; CI, Confidence Interval.

As supporting information, we present the trends calculated by Prophet models based on data from 1997 to 2018, as well as the monthly number of suicides to show suicide trends in Korea over the past two decades (S1–S3 Figs). According to the suicide trends through 2018, the number of suicides peaked in 2010–2012, and then declined gradually for both men and women (S1 and S2 Figs), and for all age groups except 35–49 years (S3 Fig).

3.2. Suicide deaths per year before and after the COVID-19 outbreak

We present the number of suicides and suicide rate per year between 2017 and 2020 (Table 4). The number of suicides and suicide rate per 100,000 population increased in 2018 and 2019 compared to 2017, but declined again in 2020. This pattern was also observed in male suicides and suicides by those aged ≥ 35 years. However, among women and those aged ≤ 34 years, suicides increased steadily from 2017, and the number and rate of suicides in 2020 also exceeded the projected estimates for that year.

Table 4

Suicides per year in Korea from 2017 to 2020.

Year	2017	2018	2019	2020
Total	12,463 (24.3)	13,670 (26.6)	13,799 (26.9) [13,548.33 (26.4)] *	13,195 (25.7) [13,000.94 (25.3)]
Men	8,922 (34.9)	9,862 (38.5)	9,730 (38.0) [9,585.87 (37.4)]	9,093 (35.5) [9,356.38 (36.5)]
Women	3,541 (13.8)	3,808 (14.8)	4,069 (15.8) [4,059.31 (15.8)]	4,102 (15.9) [3,709.20 (14.4)]
Age ≤ 34	2,150 (10.7)	2,320 (11.8)	2,380 (12.4) [2,304.41 (12.0)]	2,597 (13.8) [2,060.88 (10.9)]
Age 35–49	3,431 (27.1)	3,847 (30.7)	3,728 (30.3) [3,714.21 (30.2)]	3,470 (28.8) [3,538.26 (29.4)]
Age 50–64	3,508 (30.5)	3,910 (33.1)	4,085 (33.7) [4,015.71 (33.1)]	3,736 (30.3) [4,064.02 (32.9)]
Age ≥ 65	3,372 (47.7)	3,593 (48.6)	3,600 (46.6) [3,481.45 (45.1)]	3,392 (41.7) [3,395.28 (41.7)]

Data are shown as the number of suicides (suicides per 100,000 population).

* indicates the sum of the expected number of suicides per month according to forecasting models (with 2019 as the validation period).

4. Discussion

In this study, we determined how much the number of suicides during the pandemic differed from the expected number with the forecasting model. The forecasting models incorporated the effect of celebrity suicides on suicide trends and were validated as the expected number of suicides in 2019 closely matched the observed number in 2019. The total number of suicides per month during the COVID-19 pandemic did not deviate from the projections (Fig 1 and Table 1). However, there were differences in suicide trends among sex and age groups. During the pandemic, the monthly number of female suicides (Fig 2(B) and Table 2) and those aged ≤ 34 years (Fig 3(A) and Table 3) was significantly higher than expected.

The suicide rate in Korea fluctuates in the short and long term, although Korea has a high suicide rate. As suicide has emerged as a serious health problem in Korea, the Korean government implemented national suicide prevention policies in the early 2000s [20]. As a result, the number of suicide deaths in Korea has decreased gradually since 2011–12 [21]. Furthermore, the suicide trends are characterized by seasonal variation, with more suicides in spring and a significant decrease in winter [22]. They are also affected by copycat suicides, the phenomenon of committing suicide immediately following suicide by celebrities [23]. Considering the fluctuations in suicide trends, it might be inappropriate to compare the number of suicides during the COVID-19 pandemic directly with that of the previous years. Therefore, we first developed forecasting models based on data for the monthly number of suicides between 1997 and 2018 to predict the number of suicides in the pandemic. Then, we compared the expected and actual numbers of suicides to determine whether the COVID-19 pandemic changed suicide trends in Korea.

4.1. Suicide deaths during the COVID-19 pandemic

The COVID-19 pandemic did not affect the total number of suicides in Korea significantly, consistent with recent reports of no significant change in suicide trends after the COVID-19 outbreak in Western countries [9, 10]. In the early period of the COVID-19 pandemic, experts were concerned that suicide risk could have increased significantly due to declining incomes, social isolation, disrupted mental health delivery, and increased sales of alcohol and arms [8]. However, evidence shows that the COVID-19 pandemic has not had a significant impact on the suicide trends, at least in the early period [11]. The prevalence of COVID-19 also does not seem to correlate with an increase in suicide rates. As a plausible explanation for this, we hypothesized that the worldwide public health crisis overwhelming individuals and societies may paradoxically suppress the incidence of suicides. Emotional problems caused by personal stress can increase suicidality. Suicide rates tend to be decreased by serious external threats, such as war and natural disasters [24]. From an evolutionary perspective, the instinct to protect oneself from external danger might decrease suicidality. However, after passing through the acute crisis, suicidality can increase, along with emotional distress [25]. Therefore, we should take care because suicide rates might increase in the near future with long-lasting economic and social troubles [26].

4.2. Male and female suicides during the COVID-19 pandemic

Our results show that the COVID-19 pandemic had a greater impact on female than male suicides. The monthly number of female suicides in 2020 was significantly higher than expected (Fig 2(B) and Table 2), but the number of male suicides in 2020 did not deviate from the projection (Fig 2(A) and Table 2). In addition, the annual number of female suicides in 2020 was similar to that in 2019, and the number of male suicides in 2020 was lower than in 2019 (Table 4). In Korea, female suicides account for about one-fourth of all suicides, and were declining steadily until 2017 [21]. However, the suicide rates among women started to increase following a series of celebrity suicides at the end of 2019, and the increasing trend continued during the COVID-19 pandemic (Fig 2(B)), halting the downward trend over the past decade (S2 Fig). These findings are in line with recent studies showing that female suicides in Japan increased significantly after the first wave of the COVID-19 outbreak, suggesting that the COVID-19 pandemic had a more devastating effect on suicides among women [27, 28]. Several studies have shown that women are more vulnerable to the impact of the COVID-19 pandemic on mental health [14, 29]. Previously, we also demonstrated that women felt more psychological distress due to restrictions on social activities during the COVID-19 pandemic, while men have suffered mainly from financial difficulties [30]. Since social interaction is more crucial for coping with stress in women, prolonged social isolation and loneliness may have played an important role in the increase in suicides among women. Conversely, large-scale economic support policies for the low-income class might have prevented male suicides to an extent. Although we do not fully understand the mechanism behind the upward trend in female suicides during the pandemic, the findings suggest that there are differences in suicide risk factors in the pandemic between men and women and further studies need to address the sex difference in suicides in the COVID-19 pandemic.

4.3. Suicides by age group during the COVID-19 pandemic

The number of suicide deaths during the COVID-19 pandemic showed different patterns by age group. Compared to the projection for 2020, the monthly number of suicides by those aged ≤ 34 years was significantly higher (Fig 3(A) and Table 3), while the number of suicides by those aged 50–64 years was lower (Fig 3(C) and Table 3). In addition, compared to 2019, the annual number of suicides by those aged ≤ 34 years increased in 2020, while the number of suicides decreased in all other age groups (Table 4). In Korea, the number of suicides by those aged ≤ 34 years has been lower than that of the other age groups, and has shown a declining trend since 2011–12 (S3A Fig). By contrast, the number of suicides was highest among those aged 50–64 years in recent years, and has steadily increased in the 2010s (S3C Fig). Against this background, the different trends in suicide during the COVID-19 pandemic among age groups suggest that pandemics promote suicide among young people but suppress it among middle-aged people. Recent studies have found that young people were more vulnerable to mental health problems, including depression, during the COVID-19 pandemic [31, 32]. In addition, there have been reports of an increase in suicide attempts and suicidal thoughts among adolescents and young adults during the pandemic [33, 34]. For young people who are less vulnerable to COVID-19 infection, the restrictions in daily and economic activities due to quarantine policies might have increased their psychological burden, reflected in loneliness, depression, and despair [35-37]. They may also experience greater fear and worry about the future while living in a socioeconomically unstable society [31]. The suicide trend among young people during the COVID-19 pandemic is unclear, and further investigation of their suicidal behaviors and its risk factors is needed.

4.4. Limitations

This study had some methodological limitations. First, we used only one algorithm, Prophet, as a forecasting procedure. Although the forecasting performance of the Prophet algorithm was validated using the data for 2019 in this study, additional forecasting models are needed to find a better procedure to predict suicide trends. Second, we could not address how significantly the suicide trends changed after the COVID-19 outbreak compared to past years. As a next step, we plan to conduct an interrupted time-series analysis to verify the changes in suicide trends caused by the pandemic in Korea. Third, considering potential suicide deaths that have not been officially confirmed, the impact of the COVID-19 pandemic on suicide in Korea may have been underestimated in this study. Therefore, further research on unintentional deaths and those of undetermined intent, as well as self-harm and suicidal behaviors during the pandemic, may provide a more accurate picture [38]. Fourth, this study did not address differences in suicide incidence by educational level, economic status, or region. It is necessary to investigate suicide trends during the pandemic using an integrated model analyzing the effect of multiple risk factors, including demographic characteristics and socioeconomic conditions.

4.5. Conclusion

In conclusion, the COVID-19 pandemic has not increased the total number of suicide deaths in Korea. However, the number of female and youth suicides during the pandemic increased. This suggests that the COVID-19 pandemic might have increased the suicide risk among women and young people in Korea. Further studies of the long-term trends in suicides and related socioeconomic factors during the COVID-19 pandemic are warranted.

Performance metrics for forecasting models.

(DOCX)

Click here for additional data file.

Trends in the numbers of suicides per month between 1997 and 2018 in Korea.

The dots represent the numbers of suicides and the red dashed line indicates the trend. The blue vertical dotted lines indicate the dates of celebrity suicides.

(TIF)

Click here for additional data file.

Trends in the numbers of male and female suicides per month between 1997 and 2018 in Korea.

The dots represent the numbers of suicides and the red dashed line indicates the trend. The blue vertical dotted lines indicate the dates of celebrity suicides.

(TIF)

Click here for additional data file.

Trends in the numbers of suicides by age group per month between 1997 and 2018 in Korea.

The dots represent the numbers of suicides and the red dashed line indicates the trend. The blue vertical dotted lines indicate the dates of celebrity suicides.

(TIF)

Click here for additional data file.

3 May 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Marco Innamorati

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please upload a new copy of Figure 3 and 6 as the detail is not clear. Please follow the link for more information: https://blogs.plos.org/plos/2019/06/looking-good-tips-for-creating-your-plos-figures-graphics/

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors present sex- and age-specific suicide numbers observed in (South?) Korea between Jan 2019 and Dec 2020 and compared them with the expected values based on a forecasting procedure with data reaching back as far as 1997.

Currently we see a huge number of publications with national suicide data in relation with COVID-19. South Korea is one of the countries that deserve a somewhat higher interest given the neighborhood to Japan where suicide data seem to deviate from the world-wide development during the pandemic.

Thus, the study is generally worth to be considered for publication. However, I have two major comments.

• I think in April 2022 it is reasonable to wait for the data of 2021 and include them, too. I guess these should be available in the near future and recommend to concede the authors a sufficiently long time frame for resubmission.

• I don't see whether the Prophet tool takes the development of the general population number into account. If not so, I would strongly recommend to use suicide rates instead of numbers.

Minor points:

• Page 6, line 84 says that the blue line represents the expected numbers for 2020 but in the figures it starts already in Jan 2019.

• The legends in the figures are blurred and difficult to read.

• Figg. 4-6: why are the data of 2019 and 2020 not displayed?

• Moreover, the many dots do not contribute much to the information of importance and give a busy picture. Annually rates would in my opinion be better. Then, the necessary information could be presented as 7 lines in one figure and Figg. 5 and 6 and Table 4 could be saved.

Reviewer #2: The paper addresses the issue of putative Covid-19 pandemic mental health effects on suicide trends in Korea and concludes that there was no global increase of suicides though an increase is observable between women and young people below 34.

The introduction and review of literature is adequate and balanced and the study is highly relevant considering the high suicide rates in Korea.

The procedural rational seems theoretically correct: comparing observed suicide with forecast based on 20-years trend resorting to time series analysis, similar to ARIMA (Prophet allowing to adjust seasonality and holidays).

However, in the following papers, using joinpoint analysis, trend changes occurred in 1990 for both men and women and in 2004 for men and 2009 for women and men.

Park, C., Jee, Y. H., & Jung, K. J. (2016). Age–period–cohort analysis of the suicide rate in Korea. Journal of Affective Disorders, 194, 16-20. https://doi.org/10.1016/j.jad.2016.01.021

Lee S, Park J, Lee S, et al Changing trends in suicide rates in South Korea from 1993 to 2016: a descriptive study BMJ Open 2018;8:e023144. doi: 10.1136/bmjopen-2018-023144

Joinpoint tells us nothing on structural integrity of data and if the breakpoints are true trend changes due to preventive measures or registry errors. The present study uses trend analysis for the purpose of forecasting. Shouldn’t the quality of suicide data registry and trending between 1997 and 2019 be assessed in order to consider forecast effects?

A study using R Strucchange procedure on a national suicide long series could be useful for quality analysis:

Gusmão, R., Ramalheira, C., Conceição, V., Severo, M., Mesquita, E., Xavier, M., & Barros, H. (2021). Suicide time-series structural change analysis in Portugal (1913-2018): Impact of register bias on suicide trends. J Affect Disord, 291, 65-75. https://doi.org/10.1016/j.jad.2021.04.048

Also, on the discussion of this research we believe the ARIMA model should be included otherwise cherry-picking becomes suspicious.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

7 Jun 2022

Response to the Reviewers' comments:

Reviewer #1: The authors present sex- and age-specific suicide numbers observed in (South?) Korea between Jan 2019 and Dec 2020 and compared them with the expected values based on a forecasting procedure with data reaching back as far as 1997.

Currently we see a huge number of publications with national suicide data in relation with COVID-19. South Korea is one of the countries that deserve a somewhat higher interest given the neighborhood to Japan where suicide data seem to deviate from the world-wide development during the pandemic.

Thus, the study is generally worth to be considered for publication. However, I have two major comments.

- Response: We thank Reviewer #1 for the thoughtful suggestions.

I think in April 2022 it is reasonable to wait for the data of 2021 and include them, too. I guess these should be available in the near future and recommend to concede the authors a sufficiently long time frame for resubmission.

- Response: We examined the early impact of the COVID-19 pandemic on suicides in South Korea. The official number of suicides in 2020 was issued by Statistics Korea in October 2021. This study compared the official number of suicides in the early months of the pandemic with the expected number based on the pre-pandemic period.

We believe that data for a longer period would be needed to verify whether the suicide rate changed before and after the COVID-19 outbreak. According to Statistics Korea, the official number of suicides in 2021 should be released in the fall of 2022. As mentioned in our discussion, we plan to conduct a further study using an interrupted time-series design to detect breakpoints in suicide trends around the pandemic period.

I don't see whether the Prophet tool takes the development of the general population number into account. If not so, I would strongly recommend to use suicide rates instead of numbers.

- Response: We used Prophet models based on the monthly number of suicides without considering the total population size per month. As mentioned in the Introduction, the suicide rate in Korea has declined steadily since around 2010. Therefore, we thought that it would be inappropriate to compare the suicide rates before and after the COVID-19 outbreak directly. Instead, we compared the observed number of suicides per month in the early months of the pandemic with estimates from the Prophet model based on the pre-pandemic data. Comparing the monthly number of suicides over a year may be more sensitive than comparing monthly suicide rate per 100,000 population. Our method for comparing the observed and expected number of suicides largely followed that of Pirkis et al. (Suicide trends in the early months of the COVID-19 pandemic: an interrupted time-series analysis of preliminary data from 21 countries. Lancet Psychiatry 2021, 8(7), 579–588).

In addition, we have presented the annual number and rate of suicides in recent years in Section 3.2. and Table 4 (highlighted, page 10 line 115–121).

Minor points:

Page 6, line 84 says that the blue line represents the expected numbers for 2020 but in the figures it starts already in Jan 2019.

- Response: In Figs. 1–3, we denote the expected number of suicides for 2019 and 2020, according to forecasting models, with a blue line. First, we show the expected number for 2019 to demonstrate how closely it matches the observed number (i.e., model accuracy). The expected number for 2020 is then provided to show the extent of the deviation from the observed number of suicides for 2020. We have corrected the legend and stated that data for 2019 were used as the validation set in Figs. 1–3.

The legends in the figures are blurred and difficult to read.

- Response: We have made the lines and annotations clearer.

Fig. 4-6: why are the data of 2019 and 2020 not displayed? Moreover, the many dots do not contribute much to the information of importance and give a busy picture. Annually rates would in my opinion be better. Then, the necessary information could be presented as 7 lines in one figure and Fig. 5 and 6 and Table 4 could be saved.

- Response: Figures 4–6 show the monthly number and trend of suicides between 1997 and 2018; the forecasting models were trained on these data. In the original manuscript, we presented these data only to demonstrate that the number of suicides in Korea declined steadily from around 2010 until the COVID-19 outbreak. However, the declining trend of suicides in Korea, which has been reported in other studies, was mentioned in the Introduction of this manuscript and we do not consider these data to be the main results. Therefore, Figs. 4–6 in the original manuscript are Supplementary Figs. 1–3 in the revised manuscript (highlighted, page 6 line 86–90). Instead, following the reviewer’s recommendation, we provide the annual suicide rate (per 100,000 population) and number of suicides per year in recent years in Section 3.2 and Table 4 (highlighted, page 10 line 115–121).

Reviewer #2: The paper addresses the issue of putative Covid-19 pandemic mental health effects on suicide trends in Korea and concludes that there was no global increase of suicides though an increase is observable between women and young people below 34.

The introduction and review of literature is adequate and balanced and the study is highly relevant considering the high suicide rates in Korea.

The procedural rational seems theoretically correct: comparing observed suicide with forecast based on 20-years trend resorting to time series analysis, similar to ARIMA (Prophet allowing to adjust seasonality and holidays).

- Response: We thank Reviewer #2 for the thoughtful suggestions.

However, in the following papers, using joinpoint analysis, trend changes occurred in 1990 for both men and women and in 2004 for men and 2009 for women and men.

Park, C., Jee, Y. H., & Jung, K. J. (2016). Age–period–cohort analysis of the suicide rate in Korea. Journal of Affective Disorders, 194, 16-20. https://doi.org/10.1016/j.jad.2016.01.021

Lee S, Park J, Lee S, et al Changing trends in suicide rates in South Korea from 1993 to 2016: a descriptive study BMJ Open 2018;8:e023144. doi: 10.1136/bmjopen-2018-023144

Joinpoint tells us nothing on structural integrity of data and if the breakpoints are true trend changes due to preventive measures or registry errors. The present study uses trend analysis for the purpose of forecasting. Shouldn’t the quality of suicide data registry and trending between 1997 and 2019 be assessed in order to consider forecast effects?

A study using R Strucchange procedure on a national suicide long series could be useful for quality analysis:

Gusmão, R., Ramalheira, C., Conceição, V., Severo, M., Mesquita, E., Xavier, M., & Barros, H. (2021). Suicide time-series structural change analysis in Portugal (1913-2018): Impact of register bias on suicide trends. J Affect Disord, 291, 65-75. https://doi.org/10.1016/j.jad.2021.04.048

- Response: We appreciate the reviewer’s comments and agree fully that it is necessary to check for potential registration bias in suicide statistics in South Korea. In addition, given the possibility of underestimating the number of suicides, as mentioned in the Discussion, we need to consider presumed suicides and deaths with an undetermined cause that have not been officially reported. However, investigating these factors would require further analyses of other external cause of death statistics, or the recommended Joinpoint analysis. We consider that it is beyond the scope of this study to investigate the extent to which the number of suicides in the early months of the pandemic deviated from the projections based on the pre-pandemic period.

The suicide trends shown in Figs. 4–6, which were calculated from forecasting models trained on data from 1997 to 2018, were not the main study findings. In the original manuscript, we presented Figs. 4–6 only to demonstrate that the number of suicides in Korea had declined steadily from around 2010 until the COVID-19 outbreak (highlighted, page 6 line 86–90).

To avoid confusion, Figs. 4–6 in the original manuscript are Supplementary Figs. 1–3 in the revised manuscript. In addition, according to the reviewer’s recommendation, we plan to conduct further studies of registration bias in suicide statistics, as well as of unintentional deaths and those of underdetermined intent in Korea (highlighted, page 14 line 223–224).

Also, on the discussion of this research we believe the ARIMA model should be included otherwise cherry-picking becomes suspicious.

- Response: As shown in Figs. 1–3, there were several suicides by celebrities in Korea before and after the COVID-19 outbreak. We suspected that these celebrity suicides might have had a significant impact on suicide trends during the early stage of the pandemic. Therefore, to adjust for the effect of celebrity suicides, we used the Prophet algorithm, which has the advantage of being able to adjust for the effect of holidays and other recurring events. We have explained why we used the Prophet algorithm in Section 2.2 (highlighted, page 5 line 53–55).

In addition, following the reviewer’s recommendation, we compared the accuracy in predicting suicides in 2019 (validation period) between the Prophet and ARIMA algorithms, and found that the fitted Prophet models achieved comparable, or better, performance than the ARIMA models (highlighted, page 5 line 62–64). The RMSE, MAE, and MAPE values of the Prophet and ARIMA models calculated from the observed and expected numbers of suicides in 2019 (validation period) are provided in Supplementary Table 1.

Submitted filename: Response to reviewers.docx

Click here for additional data file.

4 Aug 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Marco Innamorati

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: Non applicable right now: the authors answers to the first review meet all my doubts and therefore I suggest immediate publication.

Reviewer #3: This is, in summary, an interesting study aimed to investigate the effect of the coronavirus disease-2019 (COVID-19) pandemic on suicide trends in Korea via a time-series analysis. The authors found that the total numbers of suicides during the COVID-19 pandemic did not deviate from projections based on the pre-pandemic period. In addition, the number of suicides among females and those under the age of 34 years significantly exceeded the expected level. Moreover, the COVID-19 pandemic did not increase the overall suicide rate significantly. Finally, suicides among women and young people increased, suggesting that the pandemic might drive more members of these groups to suicide

Overall, the present manuscript is interesting and well-written in its current version; thus, only minor changes are required, in my opinion.

The authors may find my minor comments below.

First, when throughout the Introduction section, the authors correctly stressed the psychosocial impairment and disability in the context of COVID-19, they might further stress the impact of Covid-19 related lockdown on lifestyle habits and behavioral risk factors. Importantly, the impact of COVID-19 lockdown on physical, mental, and social wellbeing of elderly and fragile populations in specific countries such as Italy cannot be ignored. The multi-disciplinary competencies together with appropriate funding and access to rich data sources may allow to fulfill interesting research objectives. Thus, according to this background, the study of Odone and coworkers published on Acta Biomed (PMID: 32701921) may be cited within the main text.

In addition, the authors might further mention, in the context of Covid-19 infection, the link between the deterioration of people’s mental health and the consequent increase in suicide rate which is frequently underreported. Unfortunately, above 2% of the traffic accidents are suicide behaviors. This phenomenon may be underreported considering that suicides by car accidents may be reported as accidental in the national statistics. Therefore, given the above information, my suggestion is to include within the manuscript, the study published in 2016 on Forensic Sci Int (PMID: 22576104).

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

8 Aug 2022

Response to the Reviewers' comments:

Reviewer #3: This is, in summary, an interesting study aimed to investigate the effect of the coronavirus disease-2019 (COVID-19) pandemic on suicide trends in Korea via a time-series analysis. The authors found that the total numbers of suicides during the COVID-19 pandemic did not deviate from projections based on the pre-pandemic period. In addition, the number of suicides among females and those under the age of 34 years significantly exceeded the expected level. Moreover, the COVID-19 pandemic did not increase the overall suicide rate significantly. Finally, suicides among women and young people increased, suggesting that the pandemic might drive more members of these groups to suicide

Overall, the present manuscript is interesting and well-written in its current version; thus, only minor changes are required, in my opinion.

- Response: We thank Reviewer #3 for the thoughtful suggestions.

First, when throughout the Introduction section, the authors correctly stressed the psychosocial impairment and disability in the context of COVID-19, they might further stress the impact of Covid-19 related lockdown on lifestyle habits and behavioral risk factors. Importantly, the impact of COVID-19 lockdown on physical, mental, and social wellbeing of elderly and fragile populations in specific countries such as Italy cannot be ignored. The multi-disciplinary competencies together with appropriate funding and access to rich data sources may allow to fulfill interesting research objectives. Thus, according to this background, the study of Odone and coworkers published on Acta Biomed (PMID: 32701921) may be cited within the main text.

- Response: Following the reviewer’s comments, we have cited the Odone et al.’s study (PMID: 32701921) in the Introduction section (highlighted, page 3 line 9 – 10).

In addition, the authors might further mention, in the context of Covid-19 infection, the link between the deterioration of people’s mental health and the consequent increase in suicide rate which is frequently underreported. Unfortunately, above 2% of the traffic accidents are suicide behaviors. This phenomenon may be underreported considering that suicides by car accidents may be reported as accidental in the national statistics. Therefore, given the above information, my suggestion is to include within the manuscript, the study published in 2016 on Forensic Sci Int (PMID: 22576104).

- Response: Thanks for the reviewer’s comments. We have added the reference (PMID: 22576104) in the Limitation section (highlighted, page 14 line 223 – 224).

Submitted filename: Response to reviewers.docx

Click here for additional data file.

12 Aug 2022

Trends in suicide deaths before and after the COVID-19 outbreak in Korea

PONE-D-22-03951R2

Dear Dr. Kim,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Marco Innamorati

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: (No Response)

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: (No Response)

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: (No Response)

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: (No Response)

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #3: In the revised paper, the authors addressed most of the major questions raised by Reviewers improving both the main structure and quality of the present manuscript. I have no further additional comments.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

**********

2 Sep 2022

PONE-D-22-03951R2

Trends in suicide deaths before and after the COVID-19 outbreak in Korea

Dear Dr. Kim:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Marco Innamorati

Academic Editor

PLOS ONE