Impact of legislation of suicide prevention act on suicide rates in the Republic of Korea: an interrupted time-series analysis

Abstract

Introduction

Suicide is a global public health problem, and the Republic of Korea has one of the highest suicide rates worldwide. In 2011, the Act on the Prevention of Suicide and the Creation of Culture of Respect for Life (Suicide Prevention Act) was introduced in the Republic of Korea to promote key initiatives for suicide prevention policies. This study aims to evaluate the effectiveness of the Act in reducing suicide rates and identify the differential effects of the legislation by age and region.

Methods

We retrieved the suicide rates by sex, age, region, and district from the Korean Statistical Information Service (KOSIS). We investigated the impact of the Suicide Prevention Act on suicide rates between 1983 and 2021. We used interrupted time-series (ITS) analysis to assess whether the trends of suicide rates changed following the introduction of the enactment and enforcement of the Suicide Prevention Act.

Results

The Suicide Prevention Act significantly reduced suicide rates in South Korea. Suicide rates had steadily increased before the Act was introduced, but after enforcement, the trend reversed with decreasing rates in both men (β3=-1.600, P-value < 0.001) and women (β3=-0.097, P-value < 0.001). In this analysis, β2 captures changes in trends before and after the Act, while β3 represents level changes following its enforcement. Age-stratified results showed the legal effectiveness in reducing suicide rates over 30 years old, but there was no change in 20–29 years old and 10–19 years old. The Act’s effectiveness in reducing suicide rates varied regionally, with the law not significantly impacting suicide rates in Ulsan (β2=-0.128, P-value = 0.944, β3=-0.380, P-value = 0.220). At the district level, around 86.2% of the regions demonstrated a positive impact, with most districts having a considerable effect and others showing a slight result on suicide prevention methods.

Conclusion

Enactment and enforcement of the Suicide Prevention Act helped to reduce suicide rates nationwide, suggesting that legal measures can be essential in implementing an integrated suicide prevention strategy. Further research should verify reasons and contexts in which the effects of suicide prevention regulations and guidelines appeared differently by age and region, as well as the mechanism of how the law could affect health indicators.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-27036-6.

Introduction

Suicide was responsible for approximately 703,000 deaths worldwide in 2019, representing about 1.3% of all deaths [1]. South Korea has the highest suicide rate among the Organisation for Economic Cooperation and Development (OECD) countries, with suicide being a major social problem [2]. The World Health Organization (WHO) recognized the importance of policies and legislation in promoting mental health and preventing suicide [3]. One of the influential strategies to improve access to mental health services and implementation research for suicide prevention could be legal frameworks. Revision of public health law supports community-based programs and public education and enhances public awareness of suicide as a public health program. Also, it helps to mandate mental health screening for individuals in schools or healthcare facilities, provide mental health services such as pharmacotherapy or counseling with preserving confidentiality and privacy of patients’ mental health condition and improve the capacity of health professionals for suicide prevention through a training program. Growing evidence supports the effectiveness of particular public health laws and policies, such as restricting access to firearms, certain medications, and other lethal means [4] and access to mental health screening and treatment services for people at risk for suicide [5].

According to the WHO 2018 report on national suicide prevention strategies, only 38 countries have implemented national suicide prevention strategies or legislation among member countries [6]. The WHO’s LIVE LIFE approach for countries to develop a comprehensive national suicide prevention strategy suggests the early identification, assessment, management, and follow-up of anyone affected by suicidal behaviors [7]. It focuses on strengthening health systems and communities to prevent suicide. One of its key components is service provision at risk of suicide through emphasizing multisectoral approaches or any initiative that includes more than one government sector, such as health, justice, law, etc [6]. WHO recommends implementing policies to increase awareness of suicide prevention and reduce access to means of suicidal behavior, such as frequently used drugs and highly lethal methods, through national and local strategies involving regulations and legislation [7]. In Japan, the Suicide Prevention Act was enacted in 2006 to promote suicide prevention strategies [8], while in the United States, the National Suicide Hotline Designation Act of 2020 was passed to establish 9-8-8 as the national suicide prevention and mental health crisis hotline number [9].

In Korea, the Act on the Prevention of Suicide and the Creation of Culture of Respect for Life (Suicide Prevention Act) was enacted on 30 March 2011 and enforced in April 2012. Since then, the state and local governments must establish and implement necessary policies for each stage, such as prevention, crisis response, and post-response, to actively rescue people at risk for suicide. The Act mandates the establishment of suicide prevention centers (Article 13) at national, provincial, and district levels, which provide early detection, counseling, education, and treatment for suicide attempters, suicide riskers, and their families. These centers are supported by allocated budgets to cover the necessary expenses for their operations. Media reporting guidelines (Article 19 − 2) were introduced to prevent the spread of suicide contagion by establishing standards for responsible reporting. This provision is particularly relevant for adolescents, who are especially vulnerable to media influence. Additionally, life-cycle-specific measures (Article 7) target different demographic groups, including adolescents, middle-aged individuals, and older adults, addressing their unique risk factors.

At the same time, it is mandated to devise necessary policies for national life-respecting culture projects, education, and public relations [10]. The Act also includes means restriction strategies, such as regulating the sale of pesticides and enhancing safety measures at high-risk locations like bridges. The Suicide Prevention Act requires comprehensive suicide prevention activities of the central and local governments and supports suicide attempters and victims and their remaining families. Its successful implementation would highly depend on the active participation of health professionals, especially in screening and preventing high-risk groups, including patients with depression and chronic physical illnesses and those with psychosomatic symptoms from various stressful life events [10].

Despite the increasing recognition of the importance of suicide prevention, the effectiveness of different suicide prevention strategies and laws remains unclear. Previous researches have identified gaps in knowledge regarding the effectiveness of legal measures in reducing suicide rates. Nakanishi using interrupted time-series (ITS) analysis (2019), reported that suicide trends were not affected by the Suicide Prevention Act (2006) in Japan [11], while the onset of the 1998 economic recession was significantly associated with increased suicide mortality rates across the overall and subgroup population stratified by sex and age. However, a similar study in the United States showed no significant effect of the Great Recession on existing trajectories of suicide rates [12]. Meanwhile, according to the national suicide prevention strategies published by WHO in 2018, the Republic of Korea witnessed a decrease in suicide rates from 31.7 to 27.3 per 100,000 people in three years after the implementation of the Act, indicating that the enactment and enforcement of suicide prevention measures would have impacted on reducing national suicide rates with some years of the maturation period [7]. No prior academic research with causal inference has been conducted, and no heterogeneity analysis to assess the differential impact of suicide prevention laws by sociodemographic characteristics has also been done.

This study examined the change in suicide rates after enacting and enforcing the Suicide Prevention Act in 2011. First, we examined whether the health legislation affected the suicide rates in the Republic of Korea. Secondly, subgroup analysis was conducted by age, gender, region, and district. Lastly, further exploration was conducted in region(s) with an outlying trend(s), if any. While the COVID-19 pandemic may have influenced suicide trends, previous studies reported declines in suicide rates during the early pandemic phase in South Korea and other countries, attributed to strengthened social and governmental support systems [13, 14]. To ensure the robustness of our findings, a sensitivity analysis comparing trends up to 2019 and 2021 was conducted.

Methods

Study design

We used ITS analysis to assess whether the trends of suicide rates changed following the introduction of the enactment and enforcement of the Suicide Prevention Act in Korea.

Materials

We used data on total suicide rates from 1983 to 2021 published online by the Korean Statistical Information Service (KOSIS) based on the certification of death. These data provide the suicide-specific mortality rates by ten-year intervals in age group and gender. The data for regions and districts were analyzed from 1998 to 2021. The suicide rates of 255 districts provided by Statistics Korea that district represents the administrative units Si/Gun/Gu in Korea. We analyzed suicide rates in 253 districts, excluding two new districts for which no data were available.

Variables

From 1983 to the year 1993, Statistic Korea used the inter- and extrapolation-based projection of population based on the latest population census as the denominator for mortality rates, while Statistic Korea used the midyear population based on resident registration number as the denominator for mortality rates after 1993 [15]. The number of suicide deaths per 100,000 population was calculated using the cause of death from KOSIS. Suicide deaths were classified based on the 9th and 10th revisions of the International Classification of Diseases (ICD-9 and ICD-10), using the intentional self-harm classification (E950–959 or X60–X84) [16].

Statistical analysis

We used ITS analysis, a robust quasi-experimental design for assessing the longitudinal impact of an intervention, as it enables both random time fluctuations and the underlying trend to be accounted for in analysis [17, 18] to evaluate legislation intervention impact on suicide rates [19]. We fit the following linear regression equation:

where y is the year, time_y counts the number of years starting from the available data year, intervention_y is the dummy variable when 0 means the intervention is yet to be implemented and 1 represents the intervention is implemented, time after intervention_y counts the number of years from 2011 when the suicide prevention law was first enacted. From the right side of the equation, β₀ is the intercept, β₁ is the baseline trend before the intervention, β₂ is the very first impact of the intervention, and β₃ is the trend change after the intervention. Statistical significance was determined as p-values of < 0.05 in a two-sided manner, while all statistical analysis was performed with STATA 17.0.

To recognize the varying impacts of the suicide prevention intervention to develop effective strategies for reducing suicide rates, we categorized four groups regarding the suicide prevention effect of law enforcement. The first group, which demonstrates a reduction in both intercept and slope, is considered a highly effective suicide prevention method that significantly lowers the suicide rate promptly as well as gradually reducing as well over time. The second group demonstrates a reduction in either the intercept or slope and is considered the group with the law’s impact on suicide prevention, instantly or gradually. The third group, which illustrates no significant differences in the intercept or slope, was considered to have no effect.

We explored a region with outlying trends from others to further examine the underlying reason for the heterogeneous impact of the intervention. A sensitivity analysis was conducted using data up to 2019 to examine whether the inclusion of the pandemic years (2020–2021) influenced the results.

Ethical statement

This study was exempt from institutional review board approval because we used publicly available data containing no personal identification.

Results

The average suicide rate per 100,000 between 1983 and 2021 was 18.5 (SD = 8.5), while the suicide rates of men were higher than women. The age-stratified analysis revealed an increasing trend in suicide rates with age, with an average of 4.6 per 100,000 for teenagers, 14.8 for those in their 20s, 19.2 for individuals in their 30s, 22.4 for those in their 40s, 25.6 for those in their 50s, and 80 for individuals over 80 years old. There have been regional disparities in average suicide rates; for example, the lowest regional suicide rate was 20.2 in Seoul, and the highest rate was 34.0 in Gangwon-do province between 1998 and 2021 (Table 1).

Table 1.

General Characteristics

Categorization	Minimum	Maximum	Mean	SD
Suicide rates (N = 39, 1983–2021)
Overall	7.3	31.7	18.5	8.5
Male	10.1	43.3	25.5	11.5
Female	4.1	22.1	11.5	5.7
Age (N = 39, 1983–2021)
0–9 years old (N = 20)	0.0	0.1	0.015	0.04
10–19 years old	3.4	7.1	4.6	1.0
20–29 years old	9.8	25.3	14.8	4.5
30–39 years old	8.7	31.4	19.2	7.2
40–49 years old	9.6	34.1	22.4	8.5
50–59 years old	9.8	41.2	25.6	10.4
60–59 years old	11.7	54.6	30.5	14.0
70–79 years old	13.0	84.4	45.2	24.5
Over 80 years old (N = 29)	23.6	128.1	80.0	33.8
Region (N = 24, 1998–2021)
Seoul Special City	9.1	26.9	20.2	5.3
Busan Metropolitan City	12.6	32.9	25.3	6.0
Daegu City	10.7	29.7	23.0	5.5
Incheon Metropolitan City	13.2	32.8	24.6	5.8
Gwangju City	10.1	30.7	20.9	5.6
Daejeon City	13.1	31.1	23.6	5.4
Ulsan Metropolitan City	12.2	28.5	21.8	4.8
Sejong City (N = 10)	17.8	41.6	22.9	7.1
Gyeonggi Province	12.7	30.5	22.8	4.8
Gangwon Province	19.3	45.2	34.0	6.5
North Chungcheong Province	17.9	42.0	30.0	6.0
South Chungcheong Province	22.8	45.8	33.6	6.6
North Jeolla Province	14.4	37.3	27.1	5.9
South Jeolla Province	15.3	33.9	26.1	5.4
North Gyeongsang Province	19.4	35.4	28.0	4.3
North Gyeongsang Province	16.1	32.2	26.1	4.1
Jeju Province	16.5	32.9	26.5	4.8

The suicide rate was estimated at 3.49 per 100,000 in 1983, and the suicide rate appeared to increase significantly every year before 2011 by 0.86 per 100,000 (P-value < 0.001). From the first year of the intervention (2011), there appeared to be a significant decrease in the annual trend of suicide rate (relative to the preintervention trend) of 1.29 per year per 100,000 (P-value < 0.001) [20]. In gender-stratified analysis, suicide rates reversed direction – from increasing before introducing the Act to decreasing afterward – in both males (β3=-0.160, P-value < 0.001) and females (β3=-0.097, P-value < 0.001) (Table 2, Figure S1).

Table 2.

Change in suicide rates (per 100,000 population) before and after the implementation of the Suicide Prevention Act by gender

Suicide rates	Coefficient		SE	t-statistics	P value
Overall	β0	3.490	1.740	2.01	0.053
	β1	0.862	0.109	7.93	0.390
	β2	1.495	1.719	0.87	0.390
	β3	-1.285	0.187	-6.87	0.000
Male	β0	5.831	2.327	2.51	0.017
	β1	1.095	0.141	7.79	0.000
	β2	3.878	2.064	1.88	0.069
	β3	-1.600	0.213	-7.50	0.000
Female	β0	1.075	1.177	0.91	0.367
	β1	0.631	0.081	7.79	0.000
	β2	-0.860	1.541	-0.56	0.580
	β3	-0.097	0.187	-5.16	0.000

β₀ intercept, β₁ baseline trend before Suicide Prevention Act are introduced, β₂ intercept change after Suicide Prevention Act was introduced, β₃ trend change after Suicide Prevention Act was introduced

Age-stratified results showed the legislation impact on reducing suicide rates over 30 years old, but there was no change in 20–29 years old (β2=-0.505, P-value = 0.861, β3=-0.264, P-value = 0.525) and 10–19 years old (β2=-0.444, P-value = 0.479, β3 = 0.139, P-value = 0.100). In the 30–59 years old, there were only decreases in the slope with no change in the intercept, and in the 60s (β2=-8.190, P-value = 0.028, β3=-3.488, P-value < 0.001) and over 80 years old (β2=-42.058, P-value = 0.001, β3=-12.306, P-value < 0.001), both the slope and the intercept decreased. The suicide prevention impact of the legislation appeared when the older age groups were increasing (Fig. 1, Table S2).

Fig. 1.

Changes in suicide rate (per 100,000 population) regarding the Suicide Prevention Act (1983–2021) by age

At the regional level, the only suicide rate in Ulsan showed no statistically significant changes between before and after 2011 (β2=-0.128, P-value = 0.944, β3=-0.380, P-value = 0.220). There was heterogeneity within the province level as well as the district level, while eight provinces had a substantial enactment impact, including Gwangju, Daejeon, Gangwon-do, Chungcheongnam-do, Jeollabuk-do, and Gyeongsangnam-do (Table 3, Figure S3, Table 4). Among the 253 districts, 34 districts (13.4%) belonged to Group 1, 184 districts (72.7%) belonged to Group 2, and 31 districts (12.3%) belonged to Group 3. Four typical districts were not easy to classify into Group 2 or Group 3, which intercept changes and slope changes had opposite directions to each other (Table 4).

Table 3.

Changes in suicide rate (per 100,000 population) before and after the implementation of the Suicide Prevention Act by region

Suicide rates	Coefficient		SE	t-statistics	P value
Seoul Special City	β0	9.820	2.492	3.94	0.001
	β1	1.258	0.299	4.21	0.000
	β2	-0.763	1.645	-0.46	0.648
	β3	-1.635	0.319	-5.13	0.000
Busan Metropolitan City	β0	13.111	2.254	5.82	0.000
	β1	1.569	0.262	5.98	0.000
	β2	-3.376	1.666	-2.03	0.056
	β3	-1.863	0.294	-6.35	0.000
Daegu City	β0	11.961	2.375	5.04	0.000
	β1	1.369	0.276	4.96	0.000
	β2	-3.348	1.672	-2.00	0.059
	β3	-1.387	0.328	-4.23	0.000
Incheon Metropolitan City	β0	13.153	1.991	6.61	0.000
	β1	1.437	0.248	5.78	0.000
	β2	-0.529	1.690	-0.31	0.757
	β3	-2.103	0.291	-7.21	0.000
Gwangju City	β0	9.075	1.683	5.39	0.000
	β1	1.521	0.249	6.10	0.000
	β2	-4.709	2.091	-2.25	0.036
	β3	-1.546	0.336	-4.61	0.000
Daejeon City	β0	12.874	1.836	7.01	0.000
	β1	1.351	0.222	6.10	0.000
	β2	-4.590	2.040	-2.25	0.036
	β3	-1.182	0.316	-3.74	0.001
Ulsan Metropolitan City	β0	14.060	2.177	6.46	0.000
	β1	0.753	0.274	2.75	0.012
	β2	-0.128	1.785	-0.07	0.944
	β3	-0.380	0.200	-1.27	0.220
Gyeonggi Province (Gyeonggi-do)	β0	13.032	1.708	7.63	0.000
	β1	1.267	0.214	5.91	0.000
	β2	-1.360	1.573	-0.86	0.398
	β3	-1.817	0.266	-6.84	0.000
Gangwon Province (Gangwon-do)	β0	22.126	2.889	7.66	0.000
	β1	1.784	0.319	5.59	0.000
	β2	-4.910	2.320	-2.12	0.047
	β3	-2.790	0.436	-6.40	0.000
North Chungcheong Province (Chungcheongbuk-do)	β0	18.051	1.757	10.28	0.000
	β1	1.674	0.233	7.19	0.000
	β2	-3.632	2.386	-1.52	0.144
	β3	-2.471	0.313	-7.90	0.000
South Chungcheong Province (Chungcheongnam-do)	β0	20.199	1.686	11.98	0.000
	β1	1.927	0.230	8.39	0.000
	β2	-5.716	2.378	-2.40	0.026
	β3	-2.707	0.380	-7.13	0.000
North Jeolla Province (Jeollabuk-do)	β0	14.737	1.048	14.06	0.000
	β1	1.714	0.104	16.41	0.000
	β2	-5.482	2.471	-2.22	0.038
	β3	-2.119	0.270	-8.06	0.000
South Jeolla Province (Jeollanam-do)	β0	15.527	1.548	10.03	0.000
	β1	1.308	0.206	6.34	0.000
	β2	-0.689	1.704	-0.40	0.690
	β3	-1.803	0.322	-5.60	0.000
North Gyeongsang Province (Gyeongsangbuk-do)	β0	19.375	1.213	15.97	0.000
	β1	1.209	0.154	7.83	0.000
	β2	-3.173	1.770	-1.79	0.088
	β3	-1.667	0.289	-5.78	0.000
South Gyeongsang Province (Gyeongsangnam-do)	β0	18.329	2.316	7.91	0.000
	β1	1.440	0.253	4.52	0.000
	β2	-4.564	1.665	-2.74	0.013
	β3	-1.420	0.311	-4.57	0.000
Jeju Province (Jeju-do)	β0	19.357	3.145	6.15	0.000
	β1	0.861	0.394	2.18	0.041
	β2	-0.664	2.974	-0.22	0.826
	β3	-1.089	0.475	-2.29	0.033

β₀ intercept, β₁ baseline trend before Suicide Prevention Act are introduced, β₂ intercept change after Suicide Prevention Act was introduced, β₃ trend change after Suicide Prevention Act was introduced

Table 4.

Changes in suicide rate (per 100,000 population) before and after the implementation of the Suicide Prevention Act

Categorization		Group 1 (Significant impact on both level and trend)	Group 2 (Significant impact on level or trend)	Group 3 (No significant change)
Nation			Overall
Gender			Male, Female
Age		60s, over 80 years olds	30s, 40s, 50s, 70s	10th, 20th
Region	City (Si), Province (Do)	Gwangju City, Daejeon City, Gangwon Province, South Chungcheong Province, North Jeolla Province, South Gyeongsang Province	Seoul Special City, Busan Metropolitan City, Daegu City, Incheon Metropolitan City, Gyeonggi Province, North Chungcheong Province, South Jeolla Province, North Gyeongsang Province, Jeju Province	Ulsan Metropolitan City
District	Seoul Special City	Seocho-gu,	Jongno-gu, Jung-gu, Yongsan-gu, Seongdong-gu, Gwangjin-gu, Dongdaemun-gu, Jungnang-gu, Seongbuk-gu, Gangbuk-gu, Dobong-gu, Nowon-gu, Eunpyeong-gu, Seodaemun-gu, Mapo-gu, Yangcheon-gu, Gangseo-gu, Guro-gu, Geumcheon-gu, Yeongdeungpo-gu, Dongjak-gu, Gwanak-gu, Gangnam-gu, Songpa-gu, Gangdong-gu
	Busan Metropolitan City	Dong-gu, Dongnae-gu, Geumjeong-gu,	Jung-gu, Seo-gu, Yeongdo-gu, Busanjin-gu, Nam-gu, Buk-gu, Haeundae-gu, Saha-gu, Gangseo-gu, Yeonje-gu, Suyeong-gu, Sasang-gu, Gijang-gun
	Daegu City	Dong-gu,	Jung-gu, Seo-gu, Nam-gu, Buk-gu, Suseong-gu, Dalseo-gu, Dalseong-gun
	Incheon Metropolitan City	Ganghwa-gun	Jung-gu, Dong-gu, Nam-gu, Yeonsu-gu, Namdong-gu, Bupyeong-gu, Gyeyang-gu, Seo-gu,
	Gwangju City		Seo-gu, Nam-gu, Buk-gu, Gwangsan-gu	Dong-gu
	Daejeon City		Dong-gu, Seo-gu, Nam-gu, Buk-gu, Daedeok-gu
	Ulsan Metropolitan City		Buk-gu	Jung-gu, Nam-gu, Dong-gu, Ulju-gun
	Gyeonggi Province	Bundang-gu, Dongducheon-si, Guri-si, Namyangju-si, Yangpyeong-gun	Suwon-si, Jangan-gu, Gwonseon-gu, Paldal-gu, Yeongtong-gu, Seongnam-si, Sujeong-gu, Jungwon-gu, Uijeongbu-si, Anyang-si, Manan-gu, Dongan-gu, Bucheon-si, Wonmi-gu, Ojeong-gu, Gwangmyeong-si, Pyeongtaek-si, Ansan-si, Sangnok-gu, Danwon-gu, Goyang-si, Deogyang-gu, Osan-si, Siheung-si, Gunpo-si, Uiwang-si, Hanam-si, Yongin-si, Giheung-gu, Paju-si, Icheon-si, Anseong-si, Gwangju-si, Pocheon-si, Yeoju-gun, Yeoncheon-gun, Gapyeong-gun	Sosa-gu, Ilsanseo-gu, Gwacheon-si, Cheoin-gu, Gimpo-si, Hwaseong-si, Yangju-si,
	Gangwon Province	Chuncheon-si, Donghae-si, Hoengseong-gun, Jeongseon-gun, Goseong-gun	Wonju-si, Gangneung-si, Taebaek-si, Sokcho-si, Samcheok-si, Hongcheon-gun, Yeongwol-gun, Pyeongchang-gun, Cheorwon-gun, Hwacheon-gun, Yanggu-gun, Inje-gun	Yangyang-gun
	North Chungcheong Province	Cheongwon-gu, Jincheon-gun,	Cheongju-si, Sangdang-gu, Chungju-si, Jecheon-si, Boeun-gun, Okcheon-gun, Yeongdong-gun, Goesan-gun, Eumseong-gun, Danyang-gun	Heungdeok-gu, Jeungpyeong-gun
	South Chungcheong Province	Sebuk-gu, Gongju-si, Nonsan-si, Hongseong-gun, Taean-gun	Cheonan-si, Boryeong-si, Asan-si, Seosan-si, Geumsan-gun, Buyeo-gun, Seocheon-gun, Cheongyang-gun, Yesan-gun,	Dongnam-gu, Gyeryong-si
	North Jeolla Province	Gimje-si, Muju-gun, Jangsu-gun, Buan-gun	Jeonju-si, Wansan-gu, Deokjin-gu, Gunsan-si, Iksan-si, Jeongeup-si, Namwon-si, Wanju-gun, Jinan-gun, Imsil-gun, Sunchang-gun, Gochang-gun
	South Jeolla Province	Hwasun-gun, Jangseong-gun	Mokpo-si, Yeosu-si, Naju-si, Damyang-gun, Gwangyang-si, Gokseong-gun, Goheung-gun, Boseong-gun, Jangheung-gun, Haenam-gun, Yeongam-gun, Muan-gun, Hampyeong-gun, Yeonggwang-gun	Gangjin-gun, Wando-gun, Jindo-gun, Sinan-gun
	North Gyeongsang Province	Gyeongju-si, Gimcheon-si, Gunwi-gun,	Pohang-si, Nam-gu, Buk-gu, Andong-si, Gumi-si, Yeongju-si, Sangju-si, Mungyeong-si, Uiseong-gun, Cheongsong-gun, Yeongyang-gun, Cheongdo-gun, Seongju-gun, Yecheon-gun, Bonghwa-gun	Yeongcheon-si, Gyeongsan-si, Yeongdeok-gun, Goryeong-gun, Chilgok-gun, Uljin-gun
	South Gyeongsang Province	Jinju-si, Yangsan-si,	Tongyeong-si, Gimhae-si, Miryang-si, Geoje-si, Haman-gun, Changnyeong-gun, Goseong-gun, Namhae-gun, Hadong-gun, Sancheong-gun, Haman-gun, Geochang-gun, Hapcheon-gun	Sacheon-si, Uiryeong-gun
	Jeju Province		Jeju-si,	Seogwipo-si

Ilsandong-gu, Suji-gu, Suncheon-si, and Gurye-gun are considered atypical cases where it is difficult to classify them into Group 2 or Group 3, and the interpretation is challenging due to the opposite direction of the significant β2 and β3 coefficients

Both male and female groups belonged to Group 2, which had an instant or gradual legal intervention impact (a reduction in either intercept or slope) effect. In the age-stratified analysis, the 60s and over 80-years-olds were Group 1, which had a substantial law impact (a reduction in intercept and slope), and those in their 30s, 40s, 50s, and 70s were Group 2, which had an intervention impact. However, those in their 10s and 20s belonged to Group 3 (no reduction in both intercept and slope), which had no impact after legal intervention in 2011 (Table 4).

Ulsan province only belonged to Group 3, having no significant impact on suicide rates from 1998 to 2021 (Group 3, β2=-0.128, P-value = 0.944, β3=-0.380, P-value = 0.220). However, further period stratified analysis revealed that the early post-enactment period only until 2017 showed legislation impact that showed reversed trends from increasing to decreasing with no intercept changes (Group 2, β2 = 0.876, P-value = 0.592, β3=-0.878, P-value = 0.012) while the late-enactment period (between 2017 and 2021) showed no impact (Group 3, β2=-0.128, P-value = 0.891, β3=-0.380, P-value = 0.082) (Table 5; Fig. 2).

Table 5.

Changes in suicide rate before and after the implementation of the Suicide Prevention Act in Ulsan Metropolitan City

Suicide rates	Coefficient		SE	t-statistics	P value
Original, 1998–2021 (Group3)	β0	14.060	2.177	6.46	0.000
	β1	0.753	0.274	2.75	0.012
	β2	-0.128	1.785	-0.07	0.944
	β3	-0.380	0.200	-1.27	0.220
Until 2017, 1998–2017 (Group 2)	β0	14.060	2.222	6.33	0.000
	β1	0.753	0.380	2.69	0.016
	β2	0.986	1.805	0.55	0.592
	β3	-0.878	0.310	-2.83	0.012
1998–2021 except 2011–2017 (Group 3)	β0	14.060	1.181	11.90	0.000
	β1	0.753	0.149	5.06	0.000
	β2	-0.128	0.926	-0.14	0.891
	β3	-0.380	0.207	-1.83	0.082

Fig. 2.

Changes in suicide rate before and after the implementation of the Suicide Prevention Act in Ulsan Metropolitan City

Sensitivity analysis confirmed consistent findings between analyses using data up to 2019 and 2021. For data up to 2019, the trend change was β2 = 1.894 (P = 0.278) and the level change was β3 = -1.426 (P < 0.001). For data up to 2021, the trend change was β2 = 1.495 (P = 0.390) and the level change was β3 = -1.285 (P < 0.001). These results indicate that the COVID-19 pandemic did not significantly affect the evaluation of the Act’s impact (Table S3).

Discussion

Before the Act was implemented, suicide rates continued to increase, but after law enactment of the Suicide Prevention Act to protect citizens’ individual life by promoting necessary matters for suicide prevention policies in 2011, the trend reversed to decreasing trend in both men and women as well as most of the subnational regions. The law effectively reduced suicide rates in those over 30, except those aged 20–29 and teenagers. The findings suggest that legal measures can be a comprehensive suicide prevention strategy. However, further research is needed to understand the differential effects of the legislation across age groups and subnational regions.

For adolescents and young adults, the limited effect may be influenced by increased academic pressures and stress levels, as reported by the Youth Health Behavior Online Survey conducted by the Korea Disease Control and Prevention Agency (KDCA), which highlights rising stress and depression trends in these age groups. Additionally, the media reporting guidelines under Article 19 − 2 primarily target traditional media outlets and do not fully address the digital and social media platforms widely consumed by adolescents, limiting their effectiveness. A recent study by Jang et al. (2022) found that the inclusion of media reporting guidelines in the Suicide Prevention Act effectively reduced copycat suicides following celebrity suicides among the general population [21]. However, these findings suggest the guidelines were less impactful for adolescents, who may consume different types of media and experience heightened academic and societal pressures that counteract the benefits of such interventions.

Gender differences in the effectiveness of suicide prevention measures warrant further exploration, as women may exhibit heightened sensitivity to certain psychosocial interventions. For instance, Oyama et al. (2004) highlighted that women are more likely to benefit from community-based mental health interventions due to their stronger social networks and greater willingness to express emotions, which can facilitate engagement with prevention programs [22]. Similarly, Chen et al. (2013) found that suicide trends among females were more responsive to interventions targeting environmental or social stressors [23]. However, despite these insights, the Act’s predominantly psychosocial approach may overlook materialistic factors—such as economic pressures and caregiving responsibilities—that disproportionately affect women. Addressing these materialistic factors, such as through financial aid or job security programs, may enhance the effectiveness of suicide prevention policies for women.

Regionally, disparities in suicide rate reductions reflect differences in mental health resource accessibility and socioeconomic conditions. For instance, urban regions such as Seoul benefited from concentrated mental health infrastructure, while rural and economically disadvantaged areas, such as Ulsan, faced limitations likely influenced by the shipbuilding industry crisis and limited socioeconomic interventions under the Act. The stratification by regions is further justified by Article 9 of the Act, which mandates region-specific implementation plans to address local needs effectively.

The results of this study indicate that the enactment and implementation of suicide prevention laws can significantly reduce national suicide rates. Findings from Japan provide a nuanced understanding of suicide prevention outcomes. While Nakanishi et al. (2020) reported no significant national impact of Japan’s 2006 Suicide Prevention Act on suicide trends, further studies suggest localized interventions yielded meaningful progress [11]. For example, Hasegawa et al. (2021) demonstrated substantial reductions in suicide rates linked to financial support programs targeting specific methods in regional settings [24]. Similarly, Okamura et al. (2021) highlighted gradual declines achieved through community-driven initiatives, particularly among youth populations [25]. These findings suggest that localized interventions—rather than the national legislative framework—may have driven observed successes in Japan. However, comprehensive evaluations of the legal intervention itself remain limited, and evidence supporting its effectiveness at a national level is inconclusive. In comparison, South Korea’s Suicide Prevention Act incorporated nationally mandated strategies, such as pesticide regulations and regionally tailored implementation plans, which addressed structural disparities and method-specific risks more systematically. This distinction underscores the broader impact of legislative approaches that integrate national mandates with local adaptations. By leveraging both structural and targeted strategies, South Korea’s legislation likely achieved greater overall effectiveness. This contrast highlights the importance of tailoring suicide prevention policies to the cultural, economic, and social contexts of each country. Furthermore, it underscores the need for future comparative research to distinguish between the effects of legal frameworks and localized interventions, ultimately informing the development of globally effective suicide prevention strategies.

Recent insights from Japan’s comprehensive suicide prevention policies highlight the complexity and multifaceted nature of effective suicide prevention strategies. Takeshima et al. (2015) and Takeshima (2024) delineated Japan’s suicide prevention development into distinct phases, emphasizing that legislative measures such as the Basic Act for Suicide Prevention (2006) and the General Principles of Suicide Prevention Policy (2007) significantly contributed to the development of a comprehensive and multi-sector approach to suicide prevention [26]. Additionally, Takeshima (2024) underscored the role of diverse social and health-related policies—including the Act for the Prevention of Child Abuse (2000), the Prevention of Spousal Violence Act (2001), the Basic Act on Cancer Control (2006), the revised Money Lending Business Act (2006), and the Basic Act on Measures to Prevent Damage to Health due to Alcohol (2013)—in collectively reinforcing suicide prevention efforts. These findings from Japan suggest that suicide prevention is most effective when integrated into broader social and public health frameworks, emphasizing that the observed reductions in suicide rates may not solely result from dedicated suicide prevention acts alone, but rather from synergistic interactions with various socio-economic policies and improved mental health infrastructure [26]. Thus, Korea’s future policy efforts should consider adopting similarly integrated approaches, ensuring suicide prevention measures are coordinated with comprehensive social and health policies. Considering Korea’s socioeconomic diversity and regional differences, a similar integrated policy framework—encompassing targeted economic support, mental health promotion, and broader social welfare policies alongside suicide-specific legislation—may enhance the sustainability of suicide rate reductions and address potential future fluctuations effectively.

The mechanisms through which suicide prevention laws can reduce suicide rates are complex and multifaceted [27]. Potential factors include increased awareness and education, improved access to mental health services, restricted access to lethal means, and a multisectoral approach that addresses the various factors contributing to suicide risk [3]. However, the effectiveness of suicide prevention laws is influenced by factors such as the law’s specific provisions, the context in which it is implemented, and broader social and cultural factors [28]. Further research is needed to understand these mechanisms fully and to identify the most effective strategies for suicide prevention.

This study reveals that the Suicide Prevention Act has effectively reduced suicide rates in Korea, particularly among older adults. However, the effectiveness of the Act in reducing suicide rates among younger age groups appears to be limited. One potential reason for this may be mental health deterioration among younger age groups, with an increasing number of patients diagnosed with depression and anxiety disorders significantly increasing recently, particularly among teenagers and those in their 20s [29]. Furthermore, there is a lack of dedicated suicide prevention centers for teenagers. Existing resources such as counseling centers and helplines are fragmented across multiple agencies, leading to blind spots and gaps in coverage [30]. There may be a relatively lower priority given to suicide prevention programs for young people due to the absolute difference in the number of high-risk individuals compared to the elderly population. As of 2021, the number of suicide deaths among 10-19-year-olds was 338 (7.1 deaths per 100,000 population), while for those aged 40–49 and 80 years and over, it was 2,298 (28.2 deaths per 100,000 population) and 1,239 (61.3 deaths per 100,000 population) respectively. This suggests that suicide prevention efforts might have focused more on middle-aged adults and elderly populations, with fewer resources for young people.

The limited effectiveness of suicide prevention strategies in specific groups or populations may also be due to several other reasons. Firstly, from 2011 to 2021, there has been an increase in factors that can promote suicide, including personal factors such as stress, depression, and economic hardships, as well as community factors such as traumatic events, increased celebrity suicides, economic downturn, and the COVID-19 pandemic. Secondly, the suicide prevention law enacted in 2011 may have had blind spots and gaps in coverage due to a lack of sensitivity to specific regions or age groups, such as young people, which could result in limited policy effectiveness. Thirdly, the possibility of pre-existing suicide prevention strategies already in place before the law’s enactment may also have impacted the limited effectiveness of suicide prevention policies, as evidenced by the lower suicide rates among young people before 2011 compared to another age group. Future research should consider developing targeted interventions for younger age groups and improving the accessibility and integration of suicide prevention resources to address the challenges of teenagers. Young adults face accessing mental health services.

Our analysis found that the suicide prevention law did not significantly impact suicide rates in Ulsan from 1998 to 2021, considering the entire study period. However, when we limited the analysis to the immediate post-intervention period (from 1998 to 2017 only), a significant impact of the intervention on reducing suicide rates appeared in Ulsan. It is essential, however, to consider the economic conditions in the region, such as employment and unemployment rates, which may have resulted in an underestimation of the intervention’s impact during the economic downturn from 2011 to 2017. Further research is needed to understand the unbiased effectiveness of suicide prevention strategies in Ulsan, considering regional characteristics and economic conditions. Nonetheless, our findings suggest that the intervention may have positively impacted reducing suicide rates in Ulsan, particularly during the financial downturn. The shipbuilding industry downturn in Ulsan might have had influenced a significantly negative impact on the local economy and might have led to a rise in suicide rates related to economic hardships. This can be speculated by the fact that the economic downturn in the shipbuilding industry in Ulsan led to a decline in employment rates, a rise in unemployment rates, and a decrease in economic activity [31] of which may have contributed to the higher percentage of suicides related to economic hardships in the area (Table 5, Figure S4, Figure S5) [32, 33] is might underestimate the true impact of the legislation in Ulsan, which is famous for its shipbuilding industry. The case of Ulsan highlights that in addition to public health interventions, social and financial support to alleviate economic difficulties and improve the local economy may also play a role in reducing suicide rates.

The atypical cases in some districts such as Ilsandong-gu, Suji-gu, Suncheon-si, and Gurye-gun underscore the need to monitor and evaluate suicide prevention policies to understand their effectiveness better and inform targeted interventions. The delayed impact observed in these regions highlights the complexity of factors influencing suicide rates and the potential time lag between policy implementation and its effects. Further research is necessary to elucidate the reasons and mechanisms behind these delayed impacts and to develop tailored interventions to reduce suicide rates in these areas.

While this study demonstrates the overall effectiveness of the Suicide Prevention Act in reducing suicide rates, the differential impact across subgroups and regions warrants further exploration. For adolescents and young adults, the limited effect may be tied to the media reporting guidelines under Article 19 − 2, which aim to reduce the contagion effect of suicide but appear insufficiently adapted to address the digital and social media platforms that dominate this demographic’s media consumption. Similarly, the establishment of suicide prevention centers under Article 13 has likely contributed to the pronounced reductions in urban regions, where mental health resources and services are more readily available. However, the lack of materialistic interventions in the Act, such as financial support or employment programs, may explain its limited effectiveness in rural and economically disadvantaged regions, as seen in areas like Ulsan. Older adults benefitted significantly from means restriction strategies outlined in Article 7—such as pesticide regulation and safety measures at high-risk locations—highlighting the importance of addressing method-specific risks in this demographic. These measures were particularly impactful in rural areas, where access to pesticides and high-risk locations is more prevalent. By addressing method-specific risks, the Act effectively reduced suicides among this demographic. However, further research is needed to evaluate the long-term effectiveness of these strategies and their implementation in diverse settings. In contrast, women experienced smaller reductions in suicide rates compared to men, reflecting gaps in the Act’s ability to address gender-specific stressors, such as caregiving responsibilities and economic burdens.

The Act’s effectiveness underscores the importance of psychosocial approaches in suicide prevention; however, as Mackenbach et al. emphasize, addressing health inequalities requires the integration of psychosocial strategies with materialistic support [34]. The Act’s exclusion of materialistic components, such as financial aid or job security programs, likely limited its impact on economically vulnerable populations. Moreover, the Act was implemented as a single intervention without staggered rollouts of individual provisions, limiting the ability to evaluate the specific effects of its components on different subgroups. Our study highlights the importance of evaluating subgroups; however, the lack of granular, longitudinal data precludes a comprehensive examination of the specific effects of individual components of the Act on different demographic groups. Future research should explore legislative designs that allow for more targeted assessments and incorporate subgroup analyses to better understand the law’s long-term impact. By complementing the Act’s existing provisions with materialistic support measures and tailoring interventions to the diverse needs of different demographic groups, suicide prevention policies can achieve more equitable and sustainable outcomes.

The inclusion of data from the COVID-19 pandemic period raised concerns about potential distortion of results. However, sensitivity analysis confirmed that findings were consistent regardless of whether data up to 2019 or 2021 were used. This suggests that the observed reductions in suicide rates during the pandemic likely reflect the robustness of the Act’s impact rather than pandemic-related artifacts.

There are several limitations that need to be considered while understanding the results. First, our study design was based on an observational ITS analysis. We cannot rule out the possibility of other factors contributing to the observed changes in suicide rates in case of unrealized intervention in the same period. Specifically, other concurrent social policies or socioeconomic changes during the study period may have independently influenced suicide trends, thereby potentially confounding the observed relationship between the Suicide Prevention Act and suicide rates. Additionally, it is crucial to acknowledge that suicide rates naturally fluctuate over time, often showing cyclical patterns of rapid increases followed by subsequent decreases independently of any specific interventions. Thus, the observed decline in suicide rates following the implementation of the Suicide Prevention Act may partially reflect such natural fluctuations. Therefore, careful interpretation is necessary to avoid overly attributing the observed trends exclusively to policy interventions. Second, the model’s goodness-of-fit was evaluated using the F-statistic (F [3, 35] = 76.25, P < 0.001), and confidence intervals were provided for all coefficient estimates to enhance reliability. However, residual analyses, such as tests for autocorrelation or residual normality, were not performed due to the limitations of aggregated secondary data. While these constraints may influence the interpretation of model assumptions, sensitivity analyses—including alternative lag structures and cutoff years—confirmed the robustness of the findings. These additional analyses (Supplementary Tables S2 and S3) provide further evidence supporting the validity of the applied interrupted time-series model. Third, this study defined suicide cases using ICD-10 codes X60-X84, which are widely recognized as the standard for identifying deaths due to intentional self-harm. While codes such as Y87.0 (‘Sequelae of intentional self-harm’) and Y10-Y34 (‘Undetermined intent’) were excluded, these codes often encompass deaths not definitively classified as suicide. Although this exclusion may result in undercounting, it ensures a precise focus on suicides directly attributable to intentional self-harm. This approach aligns with prior studies [35–37], ensuring consistency and comparability in the evaluation of suicide prevention policies. We acknowledge this limitation but believe it does not significantly affect the robustness of the findings, given the strong alignment with internationally accepted standards and methodologies. Fourth, we used publicly available data from KOSIS, which may have limited information on risk factors for suicide, such as mental health status and access to mental health services, which would not necessarily be evenly distributed across different subgroup populations. We could not distinguish another mental illness status, if any, but only analyzed the overall suicide rates. So we could not examine the effect of the law on specific subgroups, such as those with mental illness or those who have attempted suicide.

This study adds to the growing evidence that the enactment of suicide prevention, such as the Suicide Prevention Act, has effectively reduced suicide rates in the Republic of Korea. However, further research is needed to examine the potential impact after detailed policy adjustments and to conduct multiple-group analyses using data from other countries to understand the effectiveness of suicide prevention measures better. Investigating which intervention is better [19] targeting high-risk subgroups vs. comprehensive approaches aimed at the entire population. Therefore, future research should focus on developing more targeted interventions for high-risk subgroups and evaluating the law’s long-term impact on suicide rates.

Abbreviations

ICD

International Classification of Diseases

ITS

Interrupted Time-Series

KDCA

Korea Disease Control and Prevention Agency

KOSIS

Korean Statistical Information Service

OECD

Organisation for Economic Cooperation and Development

SD

Standard Deviation

WHO

World Health Organization

Authors’ contributions

YL and JO conceptualized and designed the study. YL performed the data analysis and drafted the initial manuscript. IK provided critical review and intellectual feedback. JO contributed to interpretation of results and manuscript editing. All authors read and approved the final manuscript.

Funding

None.

Data availability

Publicly available from KOSIS (https://kosis.kr/eng/).

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.