Prevalence of suicidal behavior in nigeria: a systematic review and meta-analysis

Abstract

Suicidal behaviors are a major public health concern, yet their burden remains poorly defined in Nigeria, a country with a predominantly young population. This study aimed to comprehensively assess the prevalence and correlates of suicidal ideation, planning, and attempt in Nigeria. We systematically searched medical literature databases for studies published up to November 2025 that reported the prevalence of suicidal behaviors in the Nigerian general population or specific subgroups, using diagnoses or validated screening tools. Random-effects meta-analysis were conducted, with heterogeneity explored using meta-regression and subgroup analysis. Overall, 53 studies comprising 132,514 individuals were included. The prevalence of suicidal ideation in the general population was 7.9% (95%CI: 4.6, 13.4), while suicidal planning and attempts were estimated at 1.9% (95%CI: 0.9, 4.0) and 1.3% (95%CI: 0.4, 4.3), respectively. Suicidal ideation was associated with sociodemographic factors, including marital status, educational attainment, and employment status, and was more common in conflict-prone regions. Compared with the general population, suicidal behaviors were more prevalent among adolescents and young people, people living with HIV, and pregnant women. Notably, the prevalence also increased over time. These findings indicate that suicidal behaviors are common in Nigeria and highlight the need for targeted intervention strategies for high-risk populations.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-36002-6.

Introduction

Mental health disorders represent a significant public health challenge^1,2, affecting one in every eight individuals³. Suicide, which accounts for over 700,000 deaths worldwide in 2021, is closely linked to underlying mental illness and represents one of the leading causes of premature mortality^4–6. Suicide deaths and suicidal behaviors are disproportionately more common among people with major depressive disorder, schizophrenia, and bipolar disorder, compared with the general population^3,5,6. The global age-standardized suicide rate was 8.9 per 100,000 people in 2021, with higher rates in males and younger people⁴. More than half of all suicides in the world (58%) occur in individuals under the age of 50, and the vast majority (88%) occur in low- and middle-income countries (LMICs)⁴. In Nigeria, one of the youngest nations globally, with a median age of about 18 years, the age-standardized suicide rate was estimated at 8.8 per 100,000 population in 2021⁴. Suicidal behaviors fall along a spectrum that includes suicidal ideation, suicidal planning, and suicidal attempts; these states represent escalating levels of risk and are important targets for early identification and prevention^7,8.

Over the past decade, global and national policy efforts have intensified in response to the rising burden of suicide. Since the launch of the WHO Mental Health Gap Action Program (mhGAP) in 2008⁹ and the release of WHO’s first global report on suicide prevention in 2014¹⁰, there has been a growing focus on suicide research, and suicide mortality has been recognized as a priority indicator within the Sustainable Development Goals (SDG3.4.2). In Nigeria, significant strides have been made in mental health policy recently. The National Mental Health Act 202¹¹, enacted in January 2023, represents a critical advancement in the country’s commitment to mental health, followed by the launch of the National Mental Health Policy and the National Suicide Prevention Strategic Framework in November 2023 by the Federal Ministry of Health^12,13. Despite these developments, research into suicide behaviors in Nigeria remains limited and fragmented, like in other developing regions, mainly due to a lack of effective and reliable vital records on suicide. Incomplete vital registration systems, underreporting due to stigma and criminalization, and variability in study design contribute to persistent knowledge gaps^14,15.

Existing literature, mainly cross-sectional studies, has focused on specific groups such as HIV patients, adolescents, the elderly, individuals with mental illness, and other sub-population groups, and reported wide-ranging estimates of suicide behavior^16–19. An older national study from 2003, which informed most estimates and policies, reported a 3.2% prevalence of suicidal ideation²⁰. However, a more recent 2015 study from a populous and diverse state found a higher rate of 7.2%²¹. These discrepancies may reflect temporal changes, regional differences, variations in study design, or differences in diagnostic tools used. Prior studies have also highlighted several determinants of suicidal behavior that may influence prevalence estimates across regions and subgroups in Nigeria. These include socio-economic instability^22–24, insurgency and armed conflict, and arising internal displacement^25,26, pervasive stigma around suicide²⁷, cultural ethos and norms regarding honor²⁸, weak health financing, limited mental health workforce capacity¹⁴, and low levels of mental health literacy^15,29. To date, the evidence remains fragmented, and no recent synthesis has integrated findings across populations, study designs, and diagnostic tools. To our knowledge, the only other review on this subject offered qualitative analysis into suicidal behavior in the general population, but did not include a meta-analysis³⁰.

Because Nigeria lacks a reliable national surveillance system for suicide, integrating evidence from both general-population studies and subgroup-specific studies offers the most comprehensive understanding of suicidal behavior and allows comparison across populations at varying levels of risk. Given the burden of suicide, recent policy attention, and the inconsistencies and gaps in existing data, an updated and comprehensive synthesis is necessary. Therefore, this study aimed to systematically review the literature and quantitatively estimate the prevalence of suicidal behavior (ideation, plan, and attempt) in Nigeria, focusing on both the general population and specific subgroups.

Results

Study selection

The study selection process is shown in Fig. 1 using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) chart³¹. A total of 53 cross-sectional studies among 132,514 individuals met the inclusion criteria: 47 reported on suicide ideation, 11 on suicide planning, and 23 on suicide attempts. Table 1 summarizes the characteristics of the studies included.

Fig. 1.

PRISMA Flow Diagram of Study Selection. The PRISMA flow diagram demonstrates the study selection process and inclusion criteria applied during screening. Inclusion criteria were as follows¹: cross-sectional studies or baseline data from cohort studies conducted in Nigeria², participants from the general population or specific sub-groups (excluding psychiatric inpatients)³, assessment of suicidal behaviour (ideation, planning, or attempt) using validated tools, standardised questionnaires, or trained specialists⁴, reporting prevalence of suicidal behaviour in Nigeria⁵, peer-reviewed original research (excluding reviews, case-control studies, case reports, non-English, and non-human studies).

Table 1.

Characteristics of studies Included.

Author (Year)	State	Setting/Population	Sample size	Mean Age, y	Female, %	Diagnostic Tool	Risk of bias
Adebowale32	Osun	Tertiary institutions	691			Suicidal Ideation Scale (adapted from Reynolds (1987)	Fair
Adewuya21	Lagos	Community	11,246	36.8	58	PHQ-9	Good
Adewuya33	Lagos	Secondary Schools	9441	15.61	50.4	MINI-Kid	Fair
Adeyemo34	Lagos	PLHIV	201	13.88	38.3	MINI-Kid	Good
Ahmed35	Ogun	Community	339	36.83	100	WHO Multi-Country Study on Women’s Health and Life Experience	Fair
Ajiboye36	Kwara	Incarcerated individuals	53	17.3	0	MINI-Kid	Fair
Akinyemi26	Ogun	Community	527	34.7	40.8	MINI	Fair
		Refugee community	444	34.7	40.8	MINI	Fair
Anene-Okeke37	Enugu	Tertiary institutions	500			Custom questionnaire	Fair
Aroyewun38	Southwest	Tertiary institutions	2702	21.24	53.1	Adult Suicidal Ideation Questionnaire (ASIQ)	Poor
Aroyewun39	Lagos	Pregnant women	412		100	Beck’s Scale for Suicide	Fair
Bamidele16	Ogun	PLHIV	412	47.4	68.4	Custom questionnaire	Good
Bankole40	Cross River	PLHIV	150	10.77	58.7	MINI-Kid	Fair
Bodeno41	Edo	Tertiary institutions	390	NA	69	Custom questionnaire	Fair
Cheng42	Oyo	Out-of-School youth	449	16.65	51	Single-Item Question Adapted from the YRBSS	Fair
Chinawa43	Enugu & Ebonyi	Secondary schools	764	15.87	35.7	Health Kids Colorado Questionnaire	Fair
Chinawa44	Enugu	Secondary schools	450	16.2	65.6	Columbia Suicide Severity Rating Scale (C-SSRS)	Good
Dade-Matthews45	Ogun	Secondary schools	225	15.3	51.2	MINI-Kid	Good
Egbe46	35 States	PLHIV	1187	39.3	66.5	WHO-CIDI 3.0	Good
Ekejiuba47	Edo	Tertiary institutions	79,813	NA	NA	Custom questionnaire	Poor
Fasesan48	Ogun	Community	501	32.68	51.7	BDI-II	Good
Gureje20	21 States	Community	6752	36.7	51	WHO-CIDI 3.0	Good
Ighaede-Edwards49	Edo	Secondary schools	621	NA	52.7	SBQ-R	Fair
Iweama36	Adamawa	Tertiary institutions	616	21.41	33.3	SBQ-R	Good
Jegede50	Edo	Secondary schools	725	15.23	43.4	Global School-based Student Health Survey (GSHS)	Fair
Kolawole51	Kwara	Tertiary institutions	236	21.2	53.4	Custom Questionnaire	Fair
Konstantopoulos52	Lagos, Cross River, Abuja	FSW	295	28.3	100	EPDS	Poor
Kukoyi53	Ekiti	Tertiary institutions	450	NA	61.3	Custom Questionnaire	Fair
Ladi-Akinyemi54	Lagos	Tertiary institutions	750	21.5	54	PHQ-9	Good
Lasis55	Kaduna	Incarcerated individuals	262	22.34	0	MINI	Fair
Lebimoyo56	Lagos	Pregnant women	116	31	100	SBQ-R	Fair
Mapayi57	Osun	Secondary schools	500	13.96	50.4	DISC-DPS	Fair
Mgbeojedo19	Enugu	Stroke survivors	75	NA	46.7	Beck Scale for Suicide Ideation (BSSI)	Good
Mohammed58	Bauchi	Secondary schools	672	15	55.4	PHQ-A	Fair
Musami59	Borno	PLHIV	160	11.4	48.1	Kiddie Schedule for Affective Disorders and Schizophrenia (K-SADS)	Fair
Nwaogu60	Abuja and Lagos	Construction workers	382		4.5	PHQ-9	Fair
Nyundo61	Oyo	Community	750	15.2	54.8	Kutcher Adolescent Depression Scale (KADS-6)	Poor
Oguche62	Bayelsa	Secondary schools	336	NA	NA	Suicide section of the Global School Health Survey	Fair
Ogunbajo63	4 states	MSM	389	29.2	0	CES-D	Fair
Ogundipe64	Lagos	PLHIV	295	37	61.0	BDI	Fair
Ojagbemi18	Yoruba Speaking	Community	2149	75.06	53.8	WHO-CIDI 3.0	Good
Okonkwo65	Lagos	FSW	224	26.3	100	MINI	Good
Olibamoyo66	Lagos	HCW	226	35.8	66.8	Attitudes Toward Suicide Scale (ATTS)	Fair
Omigbodun17	Oyo	Secondary schools	1429	14.4	49.1	Diagnostic Interview Schedule for Children	Good
Onyebueke67	Enugu	PLHIV	360	33.52	38.6	MINI	Good
Opakunle68	Osun	Secondary schools	1015	14.84	50.9	SBQ-R	Fair
Oyedun69	Gombe	PLHIV	328	41.89	67.1	MINI	Fair
Seb-Akahomen70	Edo	PLHIV	410	40.41	75.9	MINI	Good
Shofu-Akanji71	Lagos	Congestive cardiac failure patients	98	56.1	54.1	BSID	Good
Shofu-Akanji72	Lagos	Stroke survivors	89	57.8	53	BSID	Good
Sunday73	Enugu	PLHIV	288	13.86	55.6	MINI	Good
Sweetland74	Kaduna	Community	380	35.05	38.2	PRIME-MD	Poor
Uteh75	Ondo	Tertiary institutions (Medical Students)	121	21.78	47.1	BSID	Good
Uzoh76	Enugu	Tertiary institutions	118	19.68	44.9	BSID	Good

BDI: Beck’s Depression Inventory; BSID: Beck Scale for Suicidal Ideation; CIDI: Composite International Diagnostic Interview; MINI: Mini International Neuropsychiatric Interview; PHQ; Patient Health Questionnaire; PRIME-D: Primary Care Evaluation of Mental Disorders; SBQ-R: Suicidal Behaviors Questionnaire-Revised.

Suicide ideation

A total of 47 studies with 51,735 participants were analyzed to estimate the prevalence of suicidal ideation. The studies included various populations: eight studies from the general community^{18,20,21,26,35,61,74,77}, 10 studies each on people living with HIV (PLHIV)^{16,34,40,46,48,59,64,67,69,70}, secondary school students^{17,33,43–45,49,50,57,58,73}, and on tertiary institution students^{32,37,38,41,51,53,54,68,75,78}, two studies on female sex workers (FSW)^65,76 and pregnant women^39,52, and one study each on medical students⁵⁴, construction workers⁵⁶, men who have sex with men (MSM)⁶⁰, out-of-school youth⁶³, congestive cardiac failure patients⁴², a refugee population²⁶ and stroke survivors¹⁹.

Community studies

The prevalence of suicidal ideation among community-dwelling individuals^{18,20,21,26,35,61,74,77} was estimated at 7.9% (95%CI: 4.6–13.4; I² = 99%) (Fig. 2A), based on eight studies with 22,644 participants conducted between 2007 and 2024. These studies covered a wide range of regions and populations within the country, with participants ranging from small, local samples to large, multi-state surveys. The certainty of the evidence was very low – downgraded for serious imprecision, inconsistency, and risk of bias – the sampling strategy was unclear for two studies^74,77. On excluding two studies adjudged as poor quality, the prevalence was estimated at 5.9% (95%CI: 3.6–9.4); moderate certainty evidence based on six studies among 21,514 people, representing the best available evidence (Supplementary Table 2)^{18,20,21,26,35,61}. In sensitivity analysis using Hartung-Knapp adjusted random effects models, the prevalence estimates were unchanged, but the confidence intervals were wider than in the primary analysis (Prevalence = 8.0%; 95%CI: 3.9, 15.6).

Fig. 2.

Prevalence of Suicidal Ideation in Nigeria. Forest plots showing pooled prevalence estimates for suicidal ideation across included studies, stratified by community (A), secondary schools (B), tertiary institutions (C), and PLHIV (D). Confidence intervals (95%) and heterogeneity statistics (I²) are displayed.

The studies assessed suicidal ideation using diagnostic interviews such as Composite International Diagnostic Interview (CIDI) 3.0^18,20 and the Mini International Neuropsychiatric Interview (MINI)²⁶, and standardized questionnaires such as the Kutcher Adolescent Depression Scale (KADS-6)⁷⁴, the PRIME-MD Patient Health Questionnaire⁷⁷, the 9-item Patient Health Questionnaire (PHQ-9)²¹ and the WHO Multi-Country Study on Women’s Health and Life Experience survey⁶¹. The prevalence of suicidal ideation was 6.7% (95%CI: 2.9, 14.2) based on studies that used diagnostic interviews^18,20,26, and 10.1% (95%CI: 5.3, 18.4) based on studies that used standardized questionnaires^{21,35,61,74,77}, and these estimates did not meaningfully differ (p-heterogeneity = 0.31, Table 2).

Table 2.

Risk factors of suicidal Ideation.

Characteristic	Categories	Predicted prevalence (95%CI)1	p-heterogeneity
Age, %	Adolescents, < 20y	8.7% (3.6, 19.3)	0.17
	20–35y	11.6% (5.8, 21.7)
	35–49y	7.5% (4.3, 12.7)
	50–65 y	NA
	65 + y	4.6% (1.5, 13.2)
Sex, %	Male	11.1% (4.8, 23.7)	0.33
	Female	6.1% (2.9, 12.4)
Marital status, %	Not married	4.3% (1.9, 9.2)	0.006
	Married	5.9% (3.2, 10.7)
Education	Secondary or tertiary	5.3% (3.9, 7.1)	< 0.001
	Primary or none	23.5% (9.3, 47.7)
Employment, %	Employed	3.1% (1.5, 6.6)	< 0.001
	Unemployed	35.1% (5.4, 83.6)
Region, %	North-Central	NA	0.004
	North-East	12.1% (4.3, 29.4)
	North-West	29.7% (10.8, 59.8)
	South-East	3.6% (1.4, 9.1)
	South-South	4.0% (1.5, 10.2)
	South-West	7.2% (4.4, 11.6)
Conflict-prone, %	Yes	20.5% (9.6, 38.4)	< 0.001
	No	7.8% (4.7, 12.6)
Study instrument	Diagnostic interview	6.7% (3.0, 14.2)	0.31
	Standardized questionnaire	10.1% (5.3, 18.4)
	Custom questionnaire	5.3% (1.7, 15.1)

¹Values are predicted prevalence with 95% confidence interval.

Figure 3A presents a nomogram showing the yearly trends in the prevalence of suicidal behavior from community studies. The regression line suggests the proportion of individuals reporting suicidal ideation in Nigeria may have increased slightly over time, though the linear trend was not significant (p-heterogeneity = 0.42). The Baujat plot (Supplementary Fig. 1) shows the contribution of individual studies to overall heterogeneity and their influence on the summary proportion. The study by Sweetland (2019)⁷⁷ contributed the most to heterogeneity, with a strong influence on the summary proportion, though the prevalence did not change by > 20% on excluding it. The I² did not also change meaningfully.

Fig. 3.

Yearly Trends in the Prevalence of Suicidal Behavior in Nigeria. Nomogram illustrating yearly trend of suicidal ideation and suicidal attempts across included studies. Regression lines and prediction intervals are shown.

Meta-regression (Table 2) indicated that the prevalence of suicidal ideation was significantly related to the marital status (p-heterogeneity = 0.006), educational attainment (p-heterogeneity < 0.001), employment (p-heterogeneity < 0.001), region of study location (p-heterogeneity = 0.004), and whether the study was located in a conflict-prone region or not (p-heterogeneity < 0.001). At the study level, predicted prevalence of suicidal ideation was greater among predominantly married populations (Prevalence = 5.9%; 95%CI: 3.2, 10.7) than in unmarried counterparts (Prevalence = 4.3%; 95%CI: 1.9, 9.2). Higher prevalence estimates were also observed in populations with lower educational attainment, defined as primary education or none (Prevalence = 23.5%; 95%CI: 9.3, 47.7) than among better educated counterparts (Prevalence = 5.3%; 95%CI: 3.9, 10.7), in predominantly unemployed populations (Prevalence = 35.1%; 95%CI: 5.4, 83.6) than in employed counterparts (Prevalence = 3.1%; 95%CI: 1.5, 6.6). The predicted prevalence was highest in the North-West (Prevalence = 29.7%; 95%CI: 10.8, 59.8) and North-East (Prevalence = 12.1%; 95%CI: 4.3, 29.4) regions, both of which have been substantially affected by conflicts. The other demographic factors examined were not related to the predicted prevalence of suicidal ideation: age (p-heterogeneity = 0.17), sex (p-heterogeneity = 0.33), and the type of study instrument used (p-heterogeneity = 0.31).

Other study populations

The prevalence of suicidal ideation among secondary school students^{17,33,43–45,49,50,57,73} was estimated at 11.2% (95%CI: 8.1–15.3; I² = 97%) (Fig. 2B), based on ten studies with 15,842 individuals conducted between 2008 and 2025. The prevalence of suicidal ideation among tertiary school students^{32,37,38,41,51,53,54,68,75,78} was estimated at 17.0% (95%CI: 9.2–29.3; I² = 98%) (Fig. 2C), based on ten studies with 6,574 individuals conducted between 2022 and 2025. The prevalence of suicidal ideation among PLHIV^{16,34,40,46,48,59,64,67,69,70} was estimated at 11.5% (95%CI: 7.1, 18.2; I² = 96%) (Fig. 2D), based on ten studies among 3,791 individuals conducted between 2017 and 2025. Two studies reported prevalence estimates among FSW^65,76 and pregnant women^39,52 of 18.4% (95%CI: 14.9–22.4) and 11.3% (95%CI: 4.3–26.5), respectively (Table 3). Single studies assessed suicidal ideation among construction workers⁵⁶, MSM⁶⁰, out-of-school youth⁶³, refugees²⁶, heart failure patients⁴² (Table 3), with notably high prevalence among heart failure patients, refugees, MSMs, and out-of-school youths.

Table 3.

Prevalence of suicidal behaviors in population subgroups.

	Prevalence (95%CI)	Sample Size (No. of Studies)
Suicide Ideation
Community	7.9% (4.6, 13.4)	22,644 (8)
Secondary Schools	11.2 (8.1, 15.3)	15,842 (10)
Tertiary institutions	17.0% (9.2, 29.3	6,574 (10)
Medical students	12.4% (7.1, 19.6)	121 (1)
PLHIV	11.5% (7.1, 18.2)	3,791 (10)
Female Sex Workers (FSW)	18.4% (14.9, 22.4)	519 (2)
Pregnant women	11.3% (4.3, 26.5)	528 (2)
Congestive cardiac failure patients	52.0% (41.7, 62.2)	98 (1)
Construction workers	12.1% (8.8, 16.4)	382 (1)
Men who have sex with men (MSM)	21.3% (17.4, 25.7)	389 (1)
Out-of-school Youth	20.0% (16.4, 24.1)	449 (1)
Refugees	27.3% (23.2, 31.6)	444 (1)
Stroke Survivors	4.0% (0.8, 11.2)	75 (1)
Suicide Plan
Community	1.9 (0.9, 4.0)	10,152 (4)
Secondary Schools	8.3% (4.9, 14.3)	10,502 (3)
PLHIV	4.5% (2.9, 7.0)	822 (2)
Pregnant women	1.7% (0.2, 6.1)	116 (1)
Out-of-school Youth	15.1% (12.0, 18.8)	449 (1)
Suicide Attempt
Community	0.8% (0.4, 1.5)	9,240 (3)
Secondary Schools	6.3% (3.9, 9.9)	13,446 (6)
Tertiary institution students	0.3% (< 0.001, 17.6)	80,313 (2)
PLHIV	2.6% (1.4, 4.7)	2,159 (4)
Pregnant women	1.7% (0.2, 6.1)	116 (1)
Lawyers	1.0% (0.0, 5.6)	97 (1)
Heart failure	1.0% (0.0, 5.6)	98 (1)
Stroke survivors	20.2% (12.4, 30.1)	89 (1)
Prison	2.3% (0.8, 4.9)	262 (1)
Men who have sex with men (MSM)	10.3% (7.4, 13.7)	389 (1)
Out-of-school Youth	14.3% (11.2, 17.8)	449 (1)

On excluding studies that reported suicidality without specifying whether it reflected ideation or other suicidal outcomes among secondary school students⁴⁴ and tertiary university students⁴¹, the prevalence of suicidal ideation did not meaningfully change and was 11.5% (95%CI: 8.1–16.1) and 17.1% (95%CI: 8.5–31.5). The prevalence among PLHIV was slightly lower 9.0% (95%CI: 5.6–14.2) on excluding one such study.

Suicide planning

Eleven studies involving 22,041 participants were included in the analysis of suicide planning. The studies were distributed across three subgroups: four community-based studies^18,20,35,74, three studies among secondary school students^49,71,73, two studies on PLHIV^16,69, one study among pregnant women³⁹, and one study among out-of-school youth⁶³.

Community studies

The prevalence of suicidal planning in the general community was 1.9% (95%CI: 0.9–4.0%; I² = 95%; Fig. 4A), based on four studies among 10,152 participants^18,20,35,74. The certainty of the evidence was low – downgraded for imprecision and risk of bias – the sampling strategy was unclear for one study⁷⁴. On excluding this study, the prevalence was estimated at 1.3% (95%CI: 0.8–2.0; moderate certainty evidence based on three studies^18,20,35 among 9,402 people, representing the best available evidence (Supplementary Table 2).

Fig. 4.

Prevalence of Suicide Plan in Nigeria. Forest plots showing pooled prevalence of suicidal planning in the community (A) and secondary school (B) settings.

Two studies assessed suicidal planning using diagnostic interviews (CIDI-3.0) and standardized questionnaires: Beck’s Depression Inventory (BDI-II) and KADS-6. The reported prevalence estimates significantly varied by whether studies used diagnostic interviews or standardized questionnaires (p-heterogeneity < 0.0001). The prevalence of suicidal planning was 1.0% (95%CI: 0.8, 1.2) in the studies that used the CIDI-3.0^18,20 and 4.0% (2.3, 7.0) in the studies that used the standardized questionnaires^35,74.

Other study populations

The prevalence of suicide planning was 8.5% (95%CI: 7.9, 14.3; I² = 96%) based on three studies^49,71,73 among 10,502 secondary school students (Fig. 4B). The prevalence of suicidal planning among PLHIV^16,69 was estimated at 4.5% (95%CI: 2.9–7.0; I² = 48%) based on two studies among 822 individuals conducted in 2019 and 2023. Single studies assessed suicide planning among pregnant women³⁹ and out-of-school youth⁶³.

Suicide attempt

Twenty-three studies involving 107,537 participants were included in the analysis of suicidal attempts. The studies were distributed across six different subgroups: four community-based studies^18,20,61,74, four studies involving PLHIV^16,34,40,69, six studies among secondary school students^{17,49,50,58,71,73}, two studies among tertiary students^32,62, and one study each among out-of-school youth⁶³, men who have sex with men (MSM)⁶⁰, incarcerated individuals⁴⁷, pregnant women³⁹, stroke survivors⁵⁵, and congestive cardiac failure patients⁴².

Community studies

The prevalence of suicidal attempts among community-dwelling individuals was estimated at 1.3% (95%CI: 0.4–4.3; I² = 97%; Fig. 5A), based on four studies^18,20,61,74 among 9,651 participants. The certainty of the evidence was low – downgraded for imprecision and risk of bias – the sampling strategy was unclear for one study⁷⁴. On excluding one study adjudged as poor quality, the prevalence was estimated at 0.8% (95%CI: 0.4–1.5); moderate certainty evidence based on three studies^18,20,61 among 9,240 people (Supplementary Table 2). Figure 3B presents a nomogram showing the yearly trends in the prevalence of suicidal attempts from community studies. The regression line suggests the proportion of individuals reporting suicidal attempts in Nigeria may have increased over time (p-heterogeneity = 0.004).

Fig. 5.

Prevalence of Suicide Attempts in Community Settings in Nigeria. Forest plots showing pooled prevalence of suicidal attempts in the community (A) and secondary school (B) settings.

Other study populations

The prevalence of suicide attempts was 6.3% (95%CI: 3.9, 9.9; I² = 96%) based on six studies^{17,49,50,58,71,73} among 13,446 secondary school students conducted between 2008 and 2024 (Fig. 5B). The prevalence of suicidal ideation among PLHIV^16,34,40,69 was estimated at 2.6% (95%CI: 1.4–4.7; I² = 76%) based on four studies among 822 individuals conducted in 2019 and 2023. Single studies assessed suicide attempts among out-of-school youth⁶³, MSM⁶⁰, HCWs⁷², incarcerated individuals⁴⁷, pregnant women³⁹, stroke survivors⁵⁵, and congestive cardiac failure patients⁴².

Discussion

We reviewed 52 studies conducted between 2007 and 2025 to provide pooled prevalence estimates of suicidal behaviors in Nigeria for the general population and key subgroups. Included studies assessed suicidal behaviors using several approaches, most commonly standardized clinical interviews and validated questionnaires. In the general population, the pooled prevalence of suicidal ideation was 8%, while suicidal planning and attempts were 1.9% and 1.3%, respectively. Prevalence estimates for suicidal ideation and planning were consistently higher in studies using validated questionnaires than in those using clinical interviews. Higher prevalence of suicidal ideation was associated with marital status, educational attainment, employment status, study location, and residence in conflict-affected areas. Reported suicidal behaviors appeared to increase over time, with notable variation across subgroups.

Our general population estimates for suicidal ideation, planning, and attempts are broadly comparable to those reported in a meta-analysis from Ethiopia and the WHO World Mental Health (WMH) surveys, as well as a cross-national analysis from five LMICs (Ethiopia, Uganda, South Africa, India, and Nepal)^66,79,80, which relied predominantly on the CIDI for assessment. We also observed broadly similar estimates to those from national data in Germany, which used Rasch-based depression screening approaches⁸¹. In contrast, pooled estimates from studies conducted in Europe and China, which also used varied assessment methods, were generally lower than those observed in our analysis^82,83. These cross-country differences should be interpreted cautiously because prevalence estimates are sensitive to measurement approaches and differences in recall periods (e.g., past month, 12-month, lifetime)^84,85. We excluded studies of lifetime prevalence^36,71,72,86 to reduce the influence of recall errors. Even within the studies included in this review, prevalence estimates varied by diagnostic approach, although these differences were not statistically significant. Such variation is not unexpected, as questionnaire-based measures tend to be more sensitive to transient or subthreshold suicidal thoughts. In contrast, diagnostic interviews apply stricter criteria and may underestimate prevalence in community settings⁸⁴. In addition to assessment differences, macro-level factors likely contribute to variation by county, including variation in mental health service coverage and help-seeking, sociocultural norms and stigma that affect disclosure, and differences in exposure to upstream determinants such as conflict, displacement, and economic insecurity^87,88. These factors suggest that between-country contrasts may reflect both actual epidemiological variation and systematic differences in ascertainment.

Subgroup-specific prevalence estimates suggest significant variations in the prevalence of suicidal behaviors, with consistently higher prevalence among socially and clinically vulnerable groups. For suicidal ideation, the highest estimates were reported among refugees, MSM, and out-of-school youth. Elevated prevalence was also observed among secondary and tertiary school students and among PLHIV. A broadly similar pattern was seen for suicidal planning and attempts, with particularly high estimates among out-of-school youth, secondary and tertiary school students, healthcare workers, and MSM. However, the strength of evidence varied considerably across subgroups. While multiple studies and larger sample sizes informed the estimates for secondary school students, tertiary institutions, and PLHIV, several of the highest prevalence estimates, including those among refugees, MSM, and specific clinical populations, were derived from single studies with limited sample sizes. These estimates should therefore be interpreted with caution. Nonetheless, the consistent concentration of higher prevalence across marginalized and high-risk populations aligns with existing literature. It is plausibly explained by the compounded effects of social disadvantage, stigma, discrimination, and underlying disease burden, alongside higher prevalence of common and severe mental disorders that are well-established risk factors for suicidal behavior^89–93. The elevated prevalence of suicidal behavior among secondary and tertiary students likely reflects the age pattern observed globally, with higher risk concentrated in adolescence and young adulthood^2,4. The higher estimates among tertiary students may be attributable to stressors that are more pronounced at this stage, including academic stress, financial insecurity, employment uncertainty, and the transition to independent adulthood⁹⁴. It is plausible that tertiary students are more likely to recognize and report suicidal thoughts because of greater exposure to mental health information and survey contexts. Secondary school students may also be less likely to disclose in school-based settings, where parental oversight and supervision could suppress reporting.

We also examined socio-demographic and contextual factors associated with suicidal ideation. Age-specific analyses showed no statistically significant heterogeneity across age groups. However, the predicted prevalence of suicidal ideation was highest among young adults and adolescents, and lowest among individuals aged 65 years and older. This aligns with the higher prevalence observed in secondary and tertiary students, likely reflecting higher concentration in younger age groups. This pattern is consistent with global trends, where younger populations are showing a higher burden of mental health issues and suicidal behavior^4,95–97, a critical issue highlighted by the Lancet Psychiatry Commission on youth mental health⁹⁸. The Commission emphasized that the mental health of emerging adults has steadily worsened over the past two decades, with the COVID-19 pandemic and its aftermath further intensifying this crisis⁹⁸. Factors such as developmental challenges, socio-economic pressures, identity formation stress, and increased exposure to social media and digital environments have all been linked to this growing public health issue^96,99. In contrast, older adults may benefit from greater emotional resilience¹⁰⁰, companionship, and established social networks^101,102, which could contribute to their lower prevalence of suicidal behaviors.

Study location and employment status are other key factors associated with suicidal ideation in our results. The highest prevalence was observed in the northern regions, particularly the North-East and North-West. These regions suffer from high poverty rates and have been severely impacted by conflict, terrorism, and displacement; this could partly explain our result. In fact, stratification of location by conflict exposure showed substantially higher prevalence of suicidal ideation in conflict-prone settings compared with non-conflict areas. This pattern is linked to prolonged insecurity and displacement, particularly in the North-East, where Boko Haram and ISWAP insurgencies have caused widespread violence, loss of livelihoods, and trauma, and in parts of the North-West affected by banditry and farmer-herder conflict^25,103–106. These stressors intersect with high levels of poverty, youth unemployment, and forced migration, compounding psychological distress. At the same time, cultural silence around mental illness, limited availability of professional mental health services, and minimal health insurance coverage for mental health care constrain access to timely support, potentially amplifying suicide risk in affected regions¹⁵. These findings should, however, be interpreted with caution as there were few studies from each region, potentially allowing individual non-representative studies to drive the pooled prevalence estimates. Representative community-studies in conflict-prone regions may further clarify these findings and guide the design of targeted interventions.

Unemployment was associated with higher suicidal ideation in this review. Employment may be protective through financial stability, social integration, and psychological well-being¹⁰⁷, whereas unemployment can exacerbate economic stress and undermine mental health, including depression and self-esteem. In settings with limited social safety nets, such as Nigeria, the psychological and material consequences of unemployment may be particularly pronounced. No significant associations were found with sex, marital status, or education level, factors often linked to suicide risk. This may be due to the limitations of pooled studies.

The findings of this study have important implications for public health policy in Nigeria and will support the implementation of the National Suicide Prevention Strategic Framework (NSPSF) 2023–2030^11–13. Suicide prevention efforts should prioritize primary prevention strategies that address upstream risk factors, including unemployment, poverty, and conflict-related stressors. Given the elevated burden of suicidal behaviors among adolescents and students, a nationally coordinated school-based mental health program should be prioritized, with collaboration between the Ministries of Health and Education. Integrating mental health promotion, routine screening, and referral pathways within secondary and tertiary institutions would enable early identification and support at scale¹⁰⁸. Teachers, school counsellors, and other educational staff should be trained as community gatekeepers to recognize warning signs and facilitate timely referral. These approaches are consistent with global suicide prevention frameworks that emphasize strengthening protective factors and reducing exposure to social and economic risks¹⁰⁹. At the level of secondary prevention, integrating routine screening for suicidal ideation into primary health care and other frontline services, alongside strengthening the mental health workforce, is essential. Expanding task-sharing approaches, community gatekeeper training, and the use of digital mental health platforms can improve early identification and access to care, particularly in underserved and conflict-affected regions, and align with WHO-recommended strategies for suicide prevention and mental health care integration^109,110. For tertiary prevention, improving access to evidence-based treatment, strengthening referral pathways, expanding mental health coverage within health insurance schemes, and addressing structural barriers, including the criminalization of suicide attempts, are critical to reducing harm and improving outcomes¹¹¹. Legal and policy reforms that reduce stigma and promote help-seeking are increasingly recognized as integral components of effective suicide prevention strategies^112,113. These population-wide and subgroup-specific actions provide a coherent framework for translating evidence into practice within Nigeria’s national suicide prevention agenda.

This analysis offers several strengths, including the separate pooling of data for the general population and subgroups, which provided a more nuanced understanding of suicidal behaviors across diverse populations. By grouping studies based on behavior, subgroup, and setting, we ensured more accurate inferences. Additionally, the use of GRADE principles to assess the strength of evidence increased the transparency and reliability of our findings. However, there are limitations to consider when interpreting our findings. The availability and quality of studies were limited, with many subgroups represented by only a few or single studies, highlighting significant gaps for future research. The studies also ranged from small samples with < 100 participants to large studies that recruited tens of thousands, with implications for study-level precision and meta-analysis weights. Most studies were conducted in urban areas, limiting data representation, and the scarcity of older studies restricted analysis of long-term trends. Furthermore, although the included studies used validated diagnostic tools, variation across studies may have introduced inconsistencies. Finally, methodological constraints prevented us from assessing publication bias, which could have affected the pooled estimates.

Future research in Nigeria should prioritize original, nationally representative epidemiological studies using standardized diagnostic tools and leverage existing national platforms, like the Demographic and Health Surveys (DHS), by incorporating validated mental health and suicidal behavior modules to strengthen suicide surveillance in line with NSPSF research objectives. Regionally disaggregated primary studies, particularly in rural and conflict-affected areas, are needed to better characterize geographic variation in suicidal behaviors. Focused research on high-risk and marginalized populations remains essential given current evidence gaps. Longitudinal study designs are also needed to clarify temporal trends and pathways from suicidal ideation to attempts.

We systematically reviewed the published studies of the prevalence of suicidal behaviors in Nigeria, and found that suicidal behaviors are common, with the greatest burden among certain demographic subgroups – adolescents, PLHIV, refugees, FSW, MSM, and out-of-school youth. National and subnational mental health policies and interventions should account for this variation and target the groups with the highest burden. Interpretation should consider limitations of the evidence base, including uneven geographic coverage, limited data for several subgroups, heterogeneity in assessment tools, and the predominance of cross-sectional, urban-based studies.

Methods

Reporting

This review was developed according to the PRISMA guidelines³¹ and the Joanna Briggs Institute (JBI) methodology for systematic reviews on prevalence¹¹⁴. The protocol is registered with the International Prospective Register of Systematic Reviews (PROSPERO; CRD42024559090) and has been published elsewhere¹¹⁵. The current report focuses on the suicidal behavior outcomes.

Eligibility criteria

Studies were included if they focused on suicidal behavior (ideation, planning, and attempt) diagnosed by healthcare professionals or identified through validated tools, and provided prevalence estimates among individuals in Nigeria, representing the general population or specific subgroups. There were no restrictions on age, sex, or subgroup, but studies from psychiatric patients were excluded. Eligible studies were cross-sectional surveys or baseline data from cohort studies conducted in Nigeria. We also excluded studies that did not focus on prevalence, were case-control studies, reviews, case reports, duplicates, or involved non-human research. Including case-control studies could lead to biased estimates as the proportion of cases and controls is determined by design, rather than by the underlying frequency of occurrence of the condition in the population.

Information sources and search strategy

We searched medical literature databases including PUBMED/Medline (U.S. National Library of Medicine), EMBASE (Elsevier), Proquest, PsycInfo, African Index Medicus and African Journals Online (AJOL) for studies of mental disorders in Nigeria as part of a larger project. This paper reports on suicidal behaviors. Studies from the inception of the databases until November 2025 were included. Hand-searching was also done by examining the reference list of included papers and Google Scholar for additional papers. The search strings included a combination of MeSH terms (Medical Subject Headings), Emtree terms, and topics by which articles are indexed for the PubMed/Medline and Embase databases, respectively, as well as relevant text and keywords (Supplementary Table 1). The search strings were focused on the following concepts: Nigeria, mental disorder, and prevalence. Data from multi-country studies were included only if Nigeria-only data were explicitly reported.

Study selection and data extraction

The study selection involved three rounds: an initial title/abstract review to exclude non-qualifying studies, a full-text review to determine eligibility, and data extraction from the eligible studies. Two independent investigators performed the first two rounds in duplicate, with disagreements resolved by a third reviewer. Rayyan Software was used for study screening. Data extraction and risk of bias assessment were done using a custom cloud-based spreadsheet. The extracted data were organized into five sections: study identification, participant demographics, study characteristics, outcome measures, and risk of bias assessment. No form of automation was used in study selection or data extraction.

Outcome definition

The primary outcomes were the prevalence of suicidal ideation, suicide planning and suicidal attempt, calculated for each study as the number of individuals reporting the outcome divided by the total sample size. The review aimed to estimate the current burden of suicidal behaviors; therefore, we included studies reporting the 12-month prevalence of these outcomes. Studies reporting lifetime prevalence were excluded to avoid conflating cumulative exposure with recent risk.

For suicidal ideation, studies that assessed suicidal thought, suicidal ideation or broader constructs described as suicidality were eligible for inclusion in the main analysis, recognizing variability in terminology and measurement across studies. Where the term suicidality was used, it was assumed to primarily capture suicidal thoughts unless explicitly defined otherwise. To assess the robustness of this assumption, we conduct a sensitivity analysis excluding studies that reported suicidality, and compared the resulting prevalence estimates with the primary analysis.

Suicide planning and suicide attempt were defined according to each study’s operational criteria.

Risk of bias assessment

The risk of bias was assessed using the National Institute of Health Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies¹¹³. The tool assesses population clarity, minimal non-response, consistent sampling, representativeness, sample size rigor, and clear definition of the health condition.

Effect measure and data synthesis

The effect measure is the pooled prevalence of participants with the outcome. Random-effects meta-analyses models were used to synthesize data from studies reporting on the prevalence of the three types of suicidal behavior, accounting for study variability. The main summary measure was the pooled prevalence rate estimated using the logit transformation. Meta-analysis was performed in R version 4.5.1 using the metafor package version 4.8-0. Heterogeneity was assessed using the I² statistic, categorized as minimal (0–40%), moderate (> 40–60%), substantial (60–80%), or considerable (> 80%)¹¹⁶. In addition, the prediction intervals were illustrated in the forest plots. To evaluate the influence of each study on the pooled estimate and its contribution to heterogeneity for meta-analyses with ≥ 5 studies, Baujat plots were generated to depict the relationship of each study’s influence on the pooled estimate to its contribution to the overall heterogeneity - represented by the Cochran Q-statistic¹¹⁷. Influence analysis involved sequentially omitting studies to compare the pooled estimate and I² statistic and was conducted for meta-analysis with ≥ 5 studies. Meta-regression was conducted to examine variations by age group, subgroup, and study period, with trends presented using a nomogram. Publication bias was not assessed, as traditional methods, such as funnel plots and Egger’s tests, do not apply to meta-analyses of proportions¹¹⁷. Throughout the manuscript, we use the term prevalence for all descriptive estimates; the terms pooled prevalence and predicted prevalence are used only where necessary to explicitly distinguish meta-analytic summary estimates and model-based predictions, respectively.

In sensitivity analyses, the random effects meta-analyses were repeated restricting to studies adjudged as good or fair quality to obtain best-available estimates and using Hartung-Knapp adjustment.

Certainty assessment

For each pooled prevalence measure, we systematically assessed the certainty of the evidence using a modification of Cochrane’s GRADE approach^118,119. The certainty of evidence was rated down if included studies were not likely representative based on participant selection strategy (risk of bias), if the pooled estimate was imprecise (imprecision), if the point estimates and confidence intervals from each study substantially varied (inconsistency), or if the assessed population or outcome assessment differs from the population or outcome assessment of interest (indirectness). For imprecision, the following evidence thresholds based on a rule of thumb were used: 50%, 30%, 10%, 5%, 1%, 0.10, and 0%. These thresholds were used as pragmatic anchors reflecting meaningful shifts in clinical and public health interpretation of how prevalent or rare a condition is, rather than as strict decision boundaries. While a summary grade of moderate or high certainty evidence indicates we believe the true prevalence is likely to lie close to the estimated value, a low or very low certainty evidence indicates we believe the true prevalence is probably meaningfully different from the estimated prevalence¹²⁰.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

I.A. conceived and designed the study and performed the systematic search. I.A. and M.Y. developed the preliminary protocol and assembled the study team. A.K.A., I.A., D.O., T.O., S.O.O., J.M., M.A.K., M.Y., O.T., S.O., M.O.O., and E.I. performed study screening, data extraction, and quality appraisal. A.K.A. and I.A. conducted the statistical analysis. M.A.K. and A.A. provided clinical expertise. A.K.A., I.A., and M.Y. drafted the manuscript. A.K.A., I.A., D.O., T.O., S.O.O., J.M., M.A.K., M.Y., O.T., S.O., M.O.O., E.I., and A.A. contributed to the study design, drafting, and critical revision of the manuscript. I.A. and A.A. provided supervision. All authors had full access to the data, are responsible for the integrity of the study, and approved the final version of the manuscript for publication.

Funding

The authors did not receive any funding for this work.

Data availability

The authors compiled the data used in this analysis from studies identified through a systematic review of PubMed (https://pubmed.ncbi.nlm.nih.gov), Embase (https://www.embase.com), African Journals Online (https://www.ajol.info/index.php/ajol), ProQuest (https://www.proquest.com ), PsycInfo (https://www.apa.org/pubs/databases/psycinfo), African Index Medicus (https://indexmedicus.afro.who.int), and manual searches of reference lists of included articles. The data and analysis code are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.