Abstract
Introduction:
The aim of the present study is to estimate the lifetime prevalence of suicidal ideation, plans, and attempts in a regional representative sample and the association of these outcomes with sociodemographic factors, prior mental disorders, and childhood adversities.
Material and Methods:
The PEGASUS-Murcia project is a cross-sectional survey of a representative sample of adults in Murcia that is part of the WHO World Mental Health Survey Initiative. The Composite International Diagnostic Interview (CIDI 3.0) was administered face-to-face to 2,621 participants (67.4% response rate). The main outcomes were suicidal ideation, plans, and attempts. Lifetime prevalence, age of onset, and risk factors (sociodemographic variables, mental disorders, and childhood adversities) were examined using multiple discrete-time survival models.
Results:
Lifetime prevalence (SE) of suicidal ideation, plans and attempts were 8.0% (1.1), 2.1% (0.3), and 1.2% (1.1), respectively. Prevalence of any childhood adversities was 22.1% (1.3) in the total sample and, even higher, among those with suicide related outcomes (ranging between 36.8% and 53.7%). Female sex, younger age, prior (to onset of the outcome) lifetime prevalence of mood disorders, number of mental disorders, and exposure to childhood adversity were associated with significantly increased odds of suicidal ideation and plans.
Conclusions:
Lifetime prevalence estimates of suicidality are similar to those in community epidemiological surveys. Childhood adversities and mental disorders, especially mood disorders, are important risk factors for suicidality. Early detection of these adversities and disorders should be targeted in suicide prevention programs.
Keywords: Cross-sectional survey, Childhood adversities, suicidal behavior, suicidal ideation, suicide plan, suicide attempt
RESUMEN
Introducción:
El objetivo del studio es estimar la prevalencia a lo largo de la vida de la ideación, planes e intentos suicidas en una muestra regional representativa y su asociación con factores sociodemográficos, adversidades en la infancia y trastornos mentales previos.
Material y Métodos:
El proyecto PEGASUS-Murcia es un estudio transversal en una muestra representativa de los adultos de la Región de Murcia que forma parte de la WHO World Mental Health Survey Initiative. La Composite International Diagnostic Interview (CIDI 3.0) fue administrada cara-a-cara a 2621 participantes (tasa de respuesta: 67.4%). Los resultados principales fueron la ideación, planes e intentos suicidas. La prevalencia-vida, la edad de inicio y los factores de riesgo (variables sociodemográficas, trastornos mentales y adversidades en la infancia) fueron evaluados utilizando modelos de supervivencia multivariados de tiempo discreto.
Resultados:
La prevalencia-vida (ES) de la ideación, planes e intentos suicidas fueron 8.0% (1.1), 2.1% (0.3) y 1.2% (1.1), respectivamente. La prevalencia de cualquier adversidad en la infancia fue 22.1% (1.3) en la muestra total e, incluso mayor, entre aquellos con resultados relacionados con el suicidio (rango entre 36.8 y 53.7%). El sexo femenino, jóvenes, una historia previa de trastornos afectivos antes del inicio de los resultados suicidas, el número de trastornos mentales y la exposición a adversidades en la infancia se asociaron a un incremento significativo del riesgo de ideación y planificación suicida.
Conclusiones:
Las estimaciones de la prevalencia-vida del comportamiento suicida son similares a otros estudios epidemiológicos comunitarios. Las adversidades en la infancia y los trastornos mentales, especialmente los afectivos, son importantes factores de riesgo para la suicidabilidad. La detección temprana de estas adversidades y trastornos mentales deberían contemplarse n los programas de prevención del suicidio.
Palabras clave: Estudio transversal comunitario, adversidades en la infancia, comportamiento suicida, ideación suicida, plan de suicidio, intento suicida
INTRODUCTION
Suicide is one of the leading causes of death worldwide. Although the prevalence and characteristics of suicidal thoughts and behavior vary widely over time and between regions of the world, more than 800,000 suicide deaths are reported each year, which represents an global age-standardized suicide rate of 11.4 per 100,000 person-years (15.0 for males and 8.0 for females) 1. In Spain, age-adjusted rates of suicide vary from 6.76 to 7.02 per 100,000 inhabitants depending on the restrictive vs. broad definition of suicide used 2 and time periods considered 3. Suicide attempts and ideation are relatively common in the general population 4 and a prior suicide attempt is considered the single most important predictor of subsequent suicide deaths as well as other nonfatal suicidal behaviors 1.
The WHO World Mental Health (WMH) Survey Initiative was initially designed to carry out comparable epidemiological surveys on mental disorders in different countries all over the world with similar design and diagnostic instruments 5. This initiative overcome an important factor of heterogeneity and limitation among studies focused on suicidal behavior due to different designs, populations, and instruments used. In a sample of 17 countries from different continents, the cross-national lifetime prevalence of suicidal ideation, plans, and attempts was 9.2% (standard error, SE: 0.1), 3.1% (SE: 0.1), and 2.7% (SE: 0.1), respectively 6. Data from the Spanish portion of the European Study on the Epidemiology of Mental Disorders (ESEMED) carried out between 2001–2002 yielded lower estimated (4.4% (0.3), 1.4% (0.2), and 1.5% (0.2), respectively) 7. The presence of mental disorders consistently increased the risk of suicidal behaviors 4,7–9. Different types of childhood adversities also have been associated to an increased risk for suicide ideation and attempts, especially sexual and physical abuse 10,11.
The prevalence and risk factors related to suicidal behavior may vary over time and across regions 12. Replication of prior studies with similar designs and survey instruments in different samples representative of general population is justified to overcome methodological limitations 1. For example, a nationally representative Spanish sample was surveyed between 2011–12, ten years after the ESEMED Study, in the context of the Collaborative Research on Ageing in Europe project (the COURAGE project) 8. The latter study estimated the lifetime prevalence of suicidal ideation, plans, and attempts at 3.6%, 1.92%, and 1.46% across all regions of Spain. However, substantial geographical variations has been reported among Spanish provinces with regard to suicide rates 2. National estimates should therefore be interpreted with caution and new epidemiological surveys focused on smaller regional units which independently govern mental health care and policy should be performed 13. This is the case of Autonomous Communities in Spain, as they are responsible of health care in their territories 14.
The PEGASUS-Murcia (Psychiatric Enquiry to General Population in Southeast Spain-Murcia) project is a cross-sectional survey based on a representative sample of the adult non-institutionalized general population of Murcia Region conducted between 2010 and 2012 15. It was designed to study the prevalence, protective, and risk factors of mental disorders in the context of the WMH- Survey Initiative. The aims of the current study were to estimate the lifetime prevalence of suicidal thoughts and behaviors in the general population of Murcia, and to estimate the association between suicidality, sociodemographic characteristics, mental disorders, and childhood adversities.
MATERIAL AND METHODS
This cross-sectional study has been written according the STROBE (Strengthening The Reporting of Observational Studies in Epidemiology) statement guideline 16. The protocol was approved by the Clinical Research Ethics Committee of the University Hospital Virgen de la Arrixaca of Murcia. All participating respondents provided written informed consent. No financial incentive was given for respondents’ participation.
Sampling
The PEGASUS-Murcia (“Psychiatric Enquiry to General Population in Southeast Spain-Murcia”) project is a cross-sectional population-based survey on a representative sample of the adult non-institutionalized general population of the Murcia Region (Spain) and was carried out between 2010 and 2012 as part of the World Mental Health (WMH) Survey Initiative (http://www.hcp.med.harvard.edu/wmh/) 17. The protocol with further description of the sampling frame, selection and weighting procedures and a general description of participants has been described in more details elsewhere 15,17. Briefly, a total of 2,621 participants, with an overall response rate of 67.4% (ranging from 62 to 70% across different health areas), were interviewed in person. Respondents were 50.5% male, mean age 48.6 years, 44.0% completed 12 or more years of education while 56.0% completed less, 76.2% were living in urban settings (>10,000 inhabitants) and 71.1 % were married or cohabitating.
Diagnostic assessment
The CAPI (Computer-Assisted Personal Interviewing) version of the WHO Composite International Diagnostic Interview (CIDI 3.0, hereafter referred to as CIDI), specifically adapted for use in Spain, was used to generate diagnoses based on the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition 18. The CIDI is a comprehensive, highly-structured interview applied by lay examiners in face-to-face interviews. The CIDI is divided into two parts to optimize the duration of the interview and to reduce respondent burden. Part I was administered to all respondents and Part II was only administered to a subsample of individuals consisting of all those with Part I lifetime core disorders and a probability sub-sample of the remaining Part I respondents.
The mental disorders assessed included: Mood Disorders (major depression, bipolar disorder and dysthymia), Anxiety Disorders (generalized anxiety disorder, social phobia, specific phobia, post-traumatic stress disorder, agoraphobia without panic, panic disorder, obsessive compulsive disorder and adult separation anxiety disorder), Substance Disorders (alcohol and drug abuse and/or dependence), and Impulsive Disorders (oppositional-defiant, conduct and attention deficit disorders). For substance use disorders, abuse was defined with or without dependence in recognition of abuse being a stage in the progression to dependence. Prevalence estimates of mental disorders were determined based on whether respondents’ past or current symptomatology met the lifetime diagnostic criteria for a DSM-IV disorder. Lifetime prevalence and retrospectively-reported age-of-onset (AOO) were assessed for each disorder.
Assessment of suicidal thoughts and behaviors
The suicidality module of the CIDI was used to assess the lifetime occurrence and AOO of suicide ideation, plan and attempt. The specific question asked was: “Have any of these experiences happened to you?” First the interview questioned: “You seriously thought about committing suicide”, “You made a plan for committing suicide” and after “You attempted suicide”. These 3 questions were printed in a booklet and referred to by an alphabetic letter (A, B, or C, respectively) with the aim of ensuring participants the greatest freedom of response in an effort to diminish the likelihood of the social desirability response bias frequently associated with interviewer’s administration of potentially sensitive questions. In the present study, three lifetime history suicide outcomes were considered: suicide ideation, plan and attempt considered both in the total sample and among ideators. Early onset we defined if AOO was prior to the age of 21 years, middle if AOO was between 21 and 33 years of age, and late if AOO was after 33 years of age.
Assessment of childhood adversities
Childhood adversities, defined as those occurring prior to age 18, were assessed with retrospective self-reports 19 used in a number of international general population epidemiological studies 11,20. Twelve dichotomous variables (yes/no) were investigated: four adversities related to parental maladjustment (mental illness, substance misuse, criminality and domestic violence), three adversities depicting maltreatment (physical abuse, sexual abuse, and neglect), and other adversities covering parental death, parental divorce, other parental loss, serious physical illness and family economic adversity.
Statistical analysis
Cross-tabulations were used to estimate the lifetime prevalence of suicide ideation, plans, and attempts by sex, as well as prevalence of childhood adversities among those with suicidal thoughts and behaviors. Discrete-time survival analyses with time-varying covariates using logistic regression with person-years as the unit of analysis were used to study retrospectively assessed sociodemographic (sex, age, family income, ever married, years of education and employment status), diagnostic and childhood adversities correlates of each outcome (ideation, plan, attempt, and plan and attempt among ideators) 21. Childhood adversities were assumed to have occurred prior to the outcomes, whereas the mental disorders were included as predictors only if their reported AOO was prior to the AOO of the outcome. Coefficients were converted to odds ratios (ORs) for ease of interpretation and 95% confidence intervals (95%CI) were also reported.
Weighting procedures are described in more details elsewhere 17. Briefly, part 1 weights were used to adjust for differential probabilities of selection, and an additional part 2 weight was applied to adjust for oversampling of high-risk individuals was used to restore distribution of the general population in terms of sex, age and health-care area within Murcia. All analyses were weighted with part 2 weights, and standard errors, confidence intervals and inference tests were obtained using the Taylor series linearization method 22 implemented in the STATA software Version 14.2, using the “svy” command with weights specified as “probability weights” to adjust for the effects of sampling weighting and clustering on the precision of estimates 23
RESULTS
Lifetime prevalence of suicidal behaviours
The lifetime prevalence estimates of suicidal ideation, plans, and attempts were 8.0%, 2.1%, and 1.2%, respectively (Table 1). Females had higher lifetime prevalence than males for all three outcomes. Among ideators (n = 180), the conditional probability of ever making a plan or an attempt was 26.3% and 14.3%, respectively. The conditional probability of ever making an attempt was higher among ideators with a plan (49.6%) than those without a plan (1.7%).
Table 1.
Lifetime prevalence of suicide precursors (ideation, plan and attempts) in Murcia (Spain). (Total sample n= 1459)
| Among ideators |
|||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Ideation |
Plan |
Attempt |
Plan |
Attempt |
Unplanned attempt |
Planned attempt |
|||||||||||||||
| n | % | SE | n | % | SE | n | % | SE | n | % | SE | n | % | SE | n | % | SE | n | % | SE | |
|
| |||||||||||||||||||||
| Female | 118 | 10,1 | 1,8 | 35 | 3,0 | 0,6 | 21 | 1,8 | 0,4 | 35 | 29,5 | 5,3 | 21 | 18,1 | 5,0 | 1 | 0,3 | 0,2 | 20 | 60,8 | 9,8 |
| Male | 62 | 6 | 0,8 | 17 | 1,3 | 0,3 | 8 | 0,5 | 0,2 | 17 | 21,0 | 5,0 | 8 | 8,0 | 3,6 | 2 | 3,8 | 3,3 | 6 | 23,7 | 12,5 |
| Total | 180 | 8 | 1,1 | 52 | 2,1 | 0,3 | 29 | 1,1 | 0,2 | 52 | 26,3 | 2,8 | 29 | 14,3 | 3,5 | 3 | 1,7 | 1,6 | 26 | 49,6 | 9,8 |
SE: Standar Error
Weighted Percentages
Childhood adversities
Childhood adversities were common (Table 2). The prevalence of any childhood adversity was 22.1% in the total sample and 5.5% reported two or more CAs. Prevalence was higher among individual with lifetime suicidal behaviors, ranging from 36.8% to 53.7%. The prevalence of individual childhood adversities ranged between 0.7% (sexual abuse) and 5.4% (physical abuse). The two most prevalent childhood adversities (over 10%) among suicide precursors were parental mental illness and physical abuse.
Table 2.
Prevalence of childhood adversities among participants with suicidal behaviors
| Among ideators |
||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total sample | Ideation | Plan | Attempt | Plan | Attempt | Unplanned attempt | Planned attempt | |||||||||||
|
|
|
|
|
|
|
|
|
|||||||||||
| N | % | SE | % | SE | % | SE | % | SE | % | SE | % | SE | % | SE | % | SE | ||
|
| ||||||||||||||||||
| Type of adversity | ||||||||||||||||||
| Parental mental illness | 69 | 4,7 | 0,4 | 12,3 | 2,4 | 11,1 | 5,0 | 21,1 | 8,0 | 11,1 | 5,0 | 21,1 | 8,0 | 9,1 | 11,2 | 22,3 | 8,8 | |
| Parental Substance misuse | 35 | 1,9 | 0,9 | 5,1 | 2,1 | 11,9 | 4,7 | 9,4 | 6,5 | 11,9 | 4,7 | 9,4 | 6,5 | 0,0 | . | 10,3 | 6,6 | |
| Parental criminality | 12 | 1,0 | 0,5 | 0,8 | 0,6 | 0,4 | 0,4 | 0,8 | 0,7 | 0,4 | 0,4 | 0,8 | 0,7 | 0,0 | . | 0,9 | 0,7 | |
| Family violence | 85 | 4,9 | 0,7 | 7,4 | 2,0 | 9,9 | 3,9 | 4,8 | 4,0 | 9,9 | 3,9 | 4,8 | 4,0 | 0,0 | . | 5,3 | 4,4 | |
| Physical abuse | 79 | 5,4 | 0,7 | 11,4 | 1,6 | 10,4 | 3,4 | 24,5 | 6,0 | 10,4 | 3,4 | 24,5 | 6,0 | 100,0 | . | 17,3 | 6,0 | |
| Sexual abuse | 13 | 0,7 | 0,2 | 3,5 | 1,1 | 6,0 | 3,5 | 5,3 | 5,2 | 6,0 | 3,5 | 5,3 | 5,2 | 0,0 | . | 5,8 | 5,7 | |
| Neglect | 41 | 2,8 | 0,4 | 5,5 | 1,7 | 4,6 | 3,3 | 8,5 | 5,3 | 4,6 | 3,3 | 8,5 | 5,3 | 0,0 | . | 9,3 | 6,1 | |
| Parental death | 55 | 2,8 | 0,5 | 5,6 | 2,1 | 6,4 | 4,4 | 10,0 | 7,8 | 6,4 | 4,4 | 10,0 | 7,8 | 0,0 | . | 10,9 | 8,5 | |
| Parental divorce | 14 | 1,1 | 0,4 | 1,8 | 1,5 | 4,7 | 4,2 | 3,8 | 4,4 | 4,7 | 4,2 | 3,8 | 4,4 | 0,0 | . | 4,2 | 4,9 | |
| Other parental loss | 26 | 1,5 | 0,3 | 3,1 | 1,1 | 0,0 | . | 0,0 | . | 0,0 | . | 0,0 | . | 0,0 | . | 0,0 | . | |
| Serious physical illness | 39 | 2,4 | 0,5 | 3,7 | 2,5 | 4,5 | 1,5 | 2,8 | 2,6 | 4,5 | 1,5 | 2,8 | 2,6 | 0,0 | . | 3,1 | 2,8 | |
| Family economic adversity | 13 | 0,8 | 0,3 | 0,0 | . | 0,0 | . | 0,0 | . | 0,0 | . | 0,0 | . | 0,0 | . | 0,0 | . | |
| Number of Childhood adversities | ||||||||||||||||||
| 0 | 1114 | 77,9 | 1,3 | 63,2 | 3,8 | 51,6 | 5,6 | 46,3 | 7,7 | 51,6 | 5,6 | 46,3 | 7,7 | 0,0 | . | 50,7 | 8,3 | |
| 1 | 253 | 16,6 | 1,1 | 21,7 | 4,2 | 31,1 | 6,2 | 24,0 | 10,3 | 31,1 | 6,2 | 24,0 | 10,3 | 90,9 | 11,2 | 17,6 | 9,0 | |
| 2+ | 92 | 5,5 | 1,1 | 15,1 | 2,8 | 17,2 | 5,1 | 29,7 | 10,5 | 17,2 | 5,1 | 29,7 | 10,5 | 9,1 | 11,2 | 31,7 | 10,9 | |
SE: Standar Error.
Weighted Percentages
Sociodemographic and suicide-related phenomena as risk factors
Females had significantly higher odds than males of suicidal ideation and planning, although there were no differences regarding attempts (Table 3). Younger people (aged 18–34 years) had consistently higher odds of ideation, planning, and attempts as compared to those ≥ 65 years. Participants with high-average family income had significant higher risk of suicidal planning than the reference group (low family income). There was no significant association with other sociodemographic variables. Mean (SD) age of onset of suicidal ideation, planning, and attempt was 36.9 (17), 37.1 (13.4) and 37.5 (14.6) years, respectively. Age of onset of ideation was not associated with increased risk of subsequent suicidal plan or attempt. However, the risk of experiencing suicidal planning was found to be higher within the first year after the suicidal ideation onset. Having a plan significantly increased the risk of making a suicidal attempt.
Table 3.
Multivariate logistic regression for sociodemographic risk factors for first onset of suicidal behaviors in Murcia (Spain).
| Among ideators | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||||
| Ideation * |
Plan * |
Attempt * |
Plan ** |
Attempt *** |
|||||||
| OR | 95%CI | OR | 95%CI | OR | 95%CI | OR | 95%CI | OR | 95%CI | ||
|
| |||||||||||
| Sex | |||||||||||
| Male | Ref | Ref | Ref | Ref | Ref | ||||||
| Female | 1,94 | (1.23, 3.05) | 2,51 | (1.13, 5.58) | 3,19 | (0.83, 12.32) | 1,58 | (0.58, 4.29) | 0,70 | (0.08, 5.85) | |
| Age | |||||||||||
| 18–34 | 39,07 | (20.26, 75.34) | 49,25 | (3.45, 702.57) | 63,78 | (2.24, 1812.52) | 0,95 | (0.03, 34.27) | 2,54 | (0.03, 196.07) | |
| 35–49 | 24,76 | (8.55, 71.73) | 46,18 | (2.5, 853.7) | 29,07 | (1.31, 642.87) | 2,42 | (0.04, 155.06) | 1,01 | (0.03, 39.18) | |
| 50–64 | 12,92 | (5.03, 33.24) | 26,33 | (1.85, 374.78) | 18,01 | (1.01, 322.63) | 2,70 | (0.09, 77.43) | 0,63 | (0.03, 11.73) | |
| > 65 | Ref | Ref | Ref | Ref | Ref | ||||||
| Family income † | |||||||||||
| Low | Ref | Ref | Ref | Ref | Ref | ||||||
| Low-Average | 0,87 | (0.41, 1.86) | 1,64 | (0.58, 4.62) | 0,78 | (0.24, 2.58) | 2,69 | (0.75, 9.56) | 0,05 | (0,00, 0.90) | |
| High-Average | 1,08 | (0.52, 2.23) | 2,43 | (1.23, 4.79) | 0,67 | (0.24, 1.88) | 3,45 | (1.75, 6.81) | 0,03 | (0,00, 0.66) | |
| High | 0,69 | (0.27, 1.78) | 0,80 | (0.32, 1.99) | 0,08 | (0.01, 0.86) | 1,70 | (0.3, 9.57) | 0,01 | (0,00, 0.51) | |
| Ever married | |||||||||||
| Yes | Ref | Ref | Ref | Ref | Ref | ||||||
| No | 0,99 | (0.51, 1.89) | 1,29 | (0.56, 2.98) | 0,64 | (0.11, 3.62) | 1,66 | (0.51, 5.38) | 0,06 | (0.01, 0.47) | |
| Education # | |||||||||||
| None or Primary | 2,20 | (0.66, 7.28) | 3,04 | (0.45, 20.29) | 1,14 | (0.14, 9.43) | 1,97 | (0.29, 13.53) | 0,03 | (0,00, 0.83) | |
| Basic | 2,40 | (0.99, 5.8) | 5,70 | (0.98, 32.94) | 1,57 | (0.54, 4.57) | 5,20 | (0.8, 33.75) | 0,05 | (0.01, 0.22) | |
| Secondary | 1,68 | (0.58, 4.92) | 3,83 | (0.32, 46.11) | 0,46 | (0.03, 6.92) | 3,49 | (0.25, 49.01) | 0,01 | (0,00, 0.57) | |
| College | Ref | Ref | Ref | Ref | Ref | ||||||
| Employment | |||||||||||
| Before first employment | Ref | Ref | Ref | Ref | Ref | ||||||
| After first employment | 0,61 | (0.32, 1.17) | 0,56 | (0.23, 1.37) | 0,31 | (0.08, 1.29) | 0,55 | (0.14, 2.17) | 0,12 | (0.01, 1.59) | |
| AOO (&) of ideation | |||||||||||
| Early | -- | -- | -- | 0,30 | (0.06, 1.43) | 0,31 | (0,00, 31.11) | ||||
| Mddle | -- | -- | -- | 0,92 | (0.39, 2.17) | 0,72 | (0.06, 9.07) | ||||
| Late | -- | -- | -- | Ref | Ref | ||||||
| Time since ideation onset (years) | |||||||||||
| 0 | -- | -- | -- | 26,18 | (7.48, 91.63) | 3,38 | (0.30, 37.72) | ||||
| >=1 | -- | -- | -- | Ref | Ref | ||||||
| Having a plan | |||||||||||
| No | -- | -- | -- | -- | Ref | ||||||
| Yes | -- | -- | -- | -- | 534,26 | (12.31, 23 181.29) | |||||
Family income is defined as a four-category income scale calculated as the ratio of family income in the past 12 months divided by the median income for Spain. Low income is defined as less than or equal to 0.5, low average as 0.5 to 1.0, high average as 1.0 to 2.0, and high as over 2.0;
Completed years of education (four categories: None or primary: 0–7 years; Basic: 8–11 years; Secondary: 12–15 years and College: 16 or more years of education); & AOO: Age of onset
Multivariate models for ideation, plan and attempt including sociodemographic variables;
Multivariate model including sociodemographic variables, AOO of ideation and time since ideation onset;
All variable are included in the multivariate model.
p < 0,05 (in bold text)
Childhood adversities and lifetime mental disorders as risk factors
As described in Table 4, any childhood adversity was a risk factor of suicidal planning and in those with an attempt among ideators in the multivariate model. None of the individual childhood adversities were significant except parental divorce as a risk factor of the transition from ideation to suicidal plan. The number of childhood adversities showed a consistent dose-response trend. A similar pattern was found with the precedent of any lifetime mental disorder was associated with a significant increased risk of ideation and planning, mainly due to mood disorders. The number of mental disorders was associated with ideation and suicidal planning in a dose-response relationship when adjusted to sociodemographic variables (Suppl Table 3).
Table 4.
Childhood adversities and diagnostic categories as risk factors for first onset of suicidal behaviors in Murcia (Spain).
| Among ideators | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
||||||||||||||||
| Ideation |
Plan |
Attempt |
Plan |
Attempt |
||||||||||||
| OR | 95%CI | OR | 95%CI | OR | 95%CI | OR | 95%CI | OR | 95%CI | |||||||
|
| ||||||||||||||||
| Type of childhood adversities # | ||||||||||||||||
| Parental mental illness | 1,25 | 0,45 | 3,51 | 0,68 | 0,24 | 1,90 | 1,84 | 0,45 | 7,57 | --- | --- | --- | --- | --- | --- | |
| Parental Substance misuse | 0,94 | 0,15 | 5,91 | 3,25 | 0,34 | 30,79 | 1,62 | 0,12 | 21,16 | 7,57 | 0,35 | 163,89 | --- | --- | --- | |
| Physical abuse | 0,98 | 0,26 | 3,79 | --- | --- | --- | 3,15 | 0,80 | 12,39 | --- | --- | --- | --- | --- | --- | |
| Sexual abuse | 2,07 | 0,91 | 4,73 | 1,95 | 0,42 | 9,04 | 0,46 | 0,03 | 6,47 | --- | --- | --- | --- | --- | --- | |
| Parental divorce | --- | --- | --- | --- | --- | --- | --- | --- | --- | 12,40 | 3,59 | 42,79 | --- | --- | --- | |
| Other parental loss | 0,32 | 0,09 | 1,18 | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | |
| Number of childhood adversities | ||||||||||||||||
| 0 | Ref | Ref | Ref | Ref | Ref | |||||||||||
| 1 | 1,28 | 0,60 | 2,73 | 2,41 | 1,04 | 5,58 | 1,43 | 0,14 | 14,48 | 2,28 | 0,75 | 6,90 | 1,91 | 0,28 | 12,89 | |
| 2+ | 2,53 | 1,02 | 6,27 | 2,69 | 0,63 | 11,57 | 4,02 | 0,82 | 19,77 | 0,80 | 0,21 | 3,09 | 6,28 | 1,98 | 19,89 | |
| Categories of mental disordes | ||||||||||||||||
| Any Anxiety Disorder | 0,86 | 0,34 | 2,15 | 0,62 | 0,11 | 3,51 | 0,28 | 0,03 | 2,86 | 1,14 | 0,20 | 6,54 | 0,28 | 0,01 | 6,17 | |
| Any Mood Disorder | 4,65 | 1,89 | 11,44 | 4,59 | 2,08 | 10,12 | 2,94 | 0,42 | 20,79 | 1,85 | 0,26 | 13,20 | 0,40 | 0,00 | 44,50 | |
| Any Impulse-Control Disorder | 0,46 | 0,11 | 1,91 | 0,12 | 0,00 | 3,84 | 0,14 | 0,00 | 10,67 | 0,09 | 0,00 | 30,24 | 0,27 | 0,00 | 161,49 | |
| Any Substance Use Disorder | 1,10 | 0,23 | 5,23 | 0,94 | 0,09 | 9,91 | 1,05 | 0,02 | 66,55 | 0,87 | 0,07 | 11,40 | 0,42 | 0,00 | 172,32 | |
| Number of Diagnostic categories | ||||||||||||||||
| 0 Disorders | Ref | Ref | Ref | Ref | Ref | |||||||||||
| 1 Disorder | 3,24 | 0,91 | 11,44 | 8,99 | 2,22 | 36,43 | 14,80 | 0,35 | 624,20 | 2,70 | 0,43 | 16,85 | 13,01 | 0,04 | 3770,08 | |
| 2+ Disorders | 4,77 | 0,44 | 51,30 | 12,00 | 1,36 | 105,96 | 15,09 | 0,27 | 845,78 | 2,64 | 0,11 | 65,27 | 17,72 | 0,01 | 30450,71 | |
Multivariate logistic regression with suicidal behaviors as the dependent variable and type and number of childhood adversities, number and diagnostic categories adjusted by sociodemographic variables. p < 0,05 (in bold text)
Only those type of childhood adversities with a 95%CI not including the value 1 in Suppl Table 2 were entered in the model within each suicidal behavior as the dependent variable in multivariate logistic.
DISCUSSION
Lifetime prevalence of suicidal ideation in Murcia was higher (8.0%; 95%CI: 5.8, 11.0) than previous Spanish studies (ESEMeD: 4.4%; 3.6, 5.0 7 and COURAGE: 3.7%; 2.8, 4.5 8), but similar to other European countries (7.8%; 7.3, 8.3) 24. However, lifetime prevalence of suicidal planning (2.1%; 1.5, 3.0) and attempt (1.1%; 0.8, 1.7) were similar to the Spanish studies, with overlapping confidence intervals 7,8,24 but suicidal attempt was lower than that reported internationally (2.7%; 2.6, 2.9) 6. A similar pattern has been described in both previous Spanish surveys 7,8. The PEGASUS-Murcia project was conducted between 2010–12, in the middle of an economic crisis 15, the same period of time (2011–12) as the latest Spanish survey 8 and ten years after other national and international surveys with similar methodology had been carried out 6,7,24. However, differences between Spanish provinces might partially explain the differences described for suicidal behavior, as they have been previously described with regard to suicidal rates 2.
Being female and a younger age increased the risk of suicidal behavior consistent with other national and international surveys 6–8,24. Childhood adversities increased the risk of a suicidal thoughts and behavior, even when controlling for sociodemographic variables and lifetime mental disorders. These results are consistent with prior studies 10,11,25,26. Results suggest a dose-response relationship with numbers of childhood adversities, but low frequencies of exposure to individual childhood adversities limit our statistical power to detect significant associations. It is noteworthy that, in our sample, sexual abuse was not associated with suicidality in our multivariate analysis. Similar results have been described 27,28. However, it has been consistently described as one of the most important predictor 10,11,25,29. The relatively low prevalence of sexual abuse described in our sample (0.7%) might have produced an unreliable result.
While anxiety and mood disorders were associated to all three suicidal behaviors and substance disorders to suicidal ideation in bivariate analysis, only lifetime prevalence of depressive disorders prior to age of onset of suicidal behavior was associated with suicidal ideation or planning in our sample in the multivariate model controlling for comorbidity. Furthermore, the number of diagnostic categories was significantly associated in a dose-response relationship with suicidal behavior, though only reached statistical significance for suicidal planning. Each mental disorder has been described as a significant predictor of a subsequent suicidal behavior 6–8,24, but mood disorders are considered among the strongest 9. A similar pattern has been described where association found in bivariate analyses decreased substantially in multivariate analyses controlling for comorbidity, although remained significant 30. A possible explanation for this pattern might be that comorbidity might explain some of the associations.
Similarities in the main characteristics of the design of the study (e.g. representativeness from general population, the diagnostic instrument and methods of analysis) with other national and international surveys in the context of the WMH surveys 6,7,24 enables national and international comparisons overcoming heterogeneity across studies 1,31. Nevertheless, our results should be interpreted within caution in the context of some limitations. First, PEGASUS-Murcia project obtained a response rate of 67.4%. Though not entirely satisfactory, it was above the conventionally 60% considered as a minimum standard 32, was above the overall participation rate of other European countries participating in the previous ESEMeD project 33, and participants in this survey were comparable to census data from the general population of Murcia suggesting the representativeness of the final sample 15. However, it is conceivable that some persons with mental disorders may have refused to participate to a larger extent than the general sample, so that our results may in fact underestimate the actual prevalence of suicidal behaviors in the population. Secondly, suicidal behaviors were assessed by retrospective self-report leading to a possible under-reporting and biased recall. However, several systematic reviews have shown that adults are able to recall past experiences with sufficient accuracy to provide valuable information 34,35. Thirdly, the low prevalence of suicide precursors identified may have underpowered the study to detect associated factors and interactions, so that the 95% range of the CI should have to be interpreted with caution. Fourthly, schizophrenia and other non-affective psychoses were not initially included as a diagnosis in the CIDI as they tended to be overestimated in lay-administered interviews in previous validation studies 36,37. Besides, these studies also suggest that the vast majority of respondents with clinician-diagnosed non-affective psychoses would be captured as cases because they would also meet the criteria for CIDI anxiety, mood, or substance disorders. Finally, the cross-sectional design while allowing association studies, limits the possible causal interpretation of the findings.
This survey adds new evidence highlighting the importance of a careful evaluation of mental disorders and the exposure to childhood adversities to prevent suicidal behavior. Comprehensive suicide prevention programs should not only focus on proximal risk factors, such as detection of mental disorders, but should also include the careful evaluation of distal risk factors, such as having been exposed to adversities during the childhood. The causes of a suicidal behavior are still not fully understood but it is accepted that this behavior is the result of complex interactions between numerous factors that are difficult to address in a unique study 38. Other research strategies have been suggested to go beyond the examination of sociodemographic and psychiatric risk factors, for example, in the context of the Research Domain Criteria (RDoC) framework 39, but it will take time and interdisciplinary approaches between different specialties to clarify the knowledge of the processes involved in suicide in the near future and to improve our ability to predict and prevent suicidal behavior and suicide deaths. Meanwhile additional studies are needed to identify risk and protective factors influencing such behaviors, especially if they are designed in areas where decision makers need epidemiological data to implement mental health care planning 13. This situation is particularly important in Spain 14 where there is still no national prevention program against suicide 40.
In conclusion, based on this survey conducted on a representative sample of the general population of Murcia, one of the 17 Autonomous Communities of Spain, lifetime prevalence of suicide ideation is higher than national estimates while lifetime prevalence of suicidal planning and attempt remains similar to national estimates. Childhood adversities and mental disorders, especially mood disorders, are consistent risk factors for suicidality. Population prevention programs should focus on interventions to prevent adversities during childhood, the early detection of mental disorders, especially mood disorders, and in those individuals with suicidal thoughts.
Supplementary Material
Supplement table 1. Bivariate analyses of risk factors (sociodemographic factors and mental disorders) for first onset of suicidal behaviors in Murcia (Spain).
Supplement table 3. Lifetime Mental Disorders as risk factors for first onset of suicidal behaviors in Murcia (Spain) in multivariate logistic regression models.
Supplement table 2. Bivariate analyses of Childhood adversities for first onset of suicidal behaviors in Murcia (Spain).
Acknowledgments
The authors wish to thank all participants for their collaboration. The PEGASUS-Murcia Project was supported by the Regional Health Authorities of Murcia (“Servicio Murciano de Salud and Consejería de Sanidad y Política Social”) (Decreto n° 455/2009), and the “Ayudas para proyectos de Investigación en Salud –ISCIII- del Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica” (PI12/00809). The authors thank the WMH Coordinating Center staff at Harvard and Michigan Universities for their assistance with the instrumentation, fieldwork and data analysis. These activities were supported by the United States National Institute of Mental Health (R01MH070884), the John D. and Catherine T. MacArthur Foundation, the Pfizer Foundation, the U.S. Public Health Service (R13-MH066849, R01-MH069864, and R01 DA016558), the Fogarty International Center (FIRCA R03- TW006481), the Pan American Health Organization, the Eli Lilly & Company Foundation, Ortho-McNeil Pharmaceutical, Inc., GlaxoSmithKline, Bristol-Myers Squibb and Shire. A complete list of WMH publications can be found at http://www.hcp.med.harvard.edu/wmh/.
Role of funding source
The direct and indirect funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Footnotes
Declaration of interest
RCK has served on advisory boards for Mensante Corporation, Plus One Health Management, Lake Nona Institute and US Preventive Medicine, is a co-owner of DataStat. In the past three years, RCK has been a consultant for Hoffman-La Roche, Johnson & Johnson Wellness and Prevention, and Groupe Sanofi-Aventis. There are no patents, products in development or marketed products to declare. FNM reports punctual non-financial support from Otsuka outside the submitted work in 2019.
Data sharing statement
Public access to the diagnostic instrument, including diagnostic algorithms, should be submitted to: http://www.hcp.med.harvard.edu/wmh. However, there are limitations on the availability of raw data due to ethical restrictions related to the signed consent agreements and to the signed agreement with the WHO World Mental Health Survey Initiative to limit comparative analyses to those carried out within the consortium. Requestors wishing to access a de-identified minimal dataset necessary for the purpose of monitoring our published analyses only, can apply to the principal investigator of the PEGASUS-Murcia Project, Fernando Navarro-Mateu (email: Fernando.navarro@carm.es).
References
- 1.Saxena S, Krug EG, Chestnov O, World Health Organization, editors. Preventing suicide: a global imperative. Geneva: World Health Organization, 2014. [Google Scholar]
- 2.Salmerón D, Cirera L, Ballesta M, Navarro-Mateu F. Time trends and geographical variations in mortality due to suicide and causes of undetermined intent in Spain, 1991–2008. J Public Health Oxf Engl 2013; 35: 237–45. [DOI] [PubMed] [Google Scholar]
- 3.Lopez Bernal JA, Gasparrini A, Artundo CM, McKee M. The effect of the late 2000s financial crisis on suicides in Spain: an interrupted time-series analysis. EurJ Public Health 2013; 23: 732–6. [DOI] [PubMed] [Google Scholar]
- 4.Nock MK, Borges G, Bromet EJ, Cha CB, Kessler RC, Lee S. Suicide and suicidal behavior. Epidemiol Rev 2008; 30: 133–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kessler RC, Aguilar-Gaxiola S, Alonso J, Chatterji S, Lee S, Ustun TB. The WHO World Mental Health (WMH) Surveys. Psychiatrie 2009; 6: 5–9. [PMC free article] [PubMed] [Google Scholar]
- 6.Nock MK, Borges G, Bromet EJ, et al. Cross-national prevalence and risk factors for suicidal ideation, plans and attempts. BrJPsychiatry 2008; 192: 98–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gabilondo A, Alonso J, Pinto-Meza A, et al. Prevalencia y factores de riesgo de las ideas, planes e intentos de suicidio en la población española. Resultados del estudio ESEMeD. [Prevalence and risk factors for suicide ideation, plans and attempts in the Spanish general population. Results from the ESEMeD study]. Med Clin Barc 2007; 129: 494–500. [DOI] [PubMed] [Google Scholar]
- 8.Miret M, Caballero FF, Huerta-Ramirez R, et al. Factors associated with suicidal ideation and attempts in Spain for different age groups. Prevalence before and after the onset of the economic crisis. J Affect 2014; 163: 1–9. [DOI] [PubMed] [Google Scholar]
- 9.Nock MK, Hwang I, Sampson N, et al. Cross-national analysis of the associations among mental disorders and suicidal behavior: findings from the WHO World Mental Health Surveys. PLoSMed 2009; 6: e1000123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Angelakis I, Gillespie EL, Panagioti M. Childhood maltreatment and adult suicidality: a comprehensive systematic review with meta-analysis. Psychol Med 2019; 49: 1057–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bruffaerts R, Demyttenaere K, Borges G, et al. Childhood adversities as risk factors for onset and persistence of suicidal behaviour. BrJ Psychiatry 2010; 197: 20–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Franklin JC, Ribeiro JD, Fox KR, et al. Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychol Bull 2017; 143: 187–232. [DOI] [PubMed] [Google Scholar]
- 13.Kessler RC. Psychiatric epidemiology: selected recent advances and future directions. Bull World Health Organ 2000; 78: 464–74. [PMC free article] [PubMed] [Google Scholar]
- 14.Gonzalez B, Urbanos RM, Ortega P. Oferta pública y privada de servicios sanitarios por comunidades autónomas [Public and private supply of health services by autonomous communities in Spain]. GacSanit 2004; 18 Suppl 1: 82–9. [DOI] [PubMed] [Google Scholar]
- 15.Navarro-Mateu F, Tormo MJ, Salmeron D, et al. Prevalence of Mental Disorders in the South-East of Spain, One of the European Regions Most Affected by the Economic Crisis: The Cross-Sectional PEGASUS-Murcia Project. PLOS One 2015; 10: e0137293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoSMed 2007; 4: e296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Navarro-Mateu F, Tormo M, Vilagut G, et al. Epidemiology and genetics of common mental disorders in the general population: the PEGASUS-Murcia project. BMJ Open 2013; 3: e004035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Navarro-Mateu F, Moran-Sanchez I, Alonso J, et al. Cultural adaptation of the Latin American version of the World Health Organization Composite International Diagnostic Interview (WHO-CIDI) (v 3.0) for use in Spain. GacSanit 2012; 27: 325–31. [DOI] [PubMed] [Google Scholar]
- 19.Kessler RC, McLaughlin KA, Green JG, et al. Childhood adversities and adult psychopathology in the WHO World Mental Health Surveys. Br J Psychiatry 2010; 197: 378–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Green JG, McLaughlin KA, Berglund PA, et al. Childhood adversities and adult psychiatric disorders in the national comorbidity survey replication I: associations with first onset of DSM-IV disorders. ArchGenPsychiatry 2010; 67: 113–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Efron B Logistic Regression, Survival Analysis, and the Kaplan-Meier Curve. J Am Stat Assoc 1988; 83: 414–25. [Google Scholar]
- 22.Wolter KM. Introduction to variance estimation. New York: Spinger, 1985. [Google Scholar]
- 23.StataCorp. Stata Statistical Software: Release 14. College Station, TX:StataCorp LP. 2015. [Google Scholar]
- 24.Bernal M, Haro JM, Bernert S, et al. Risk factors for suicidality in Europe: results from the ESEMED study. J Affect 2007; 101: 27–34. [DOI] [PubMed] [Google Scholar]
- 25.Liu J, Fang Y, Gong J, et al. Associations between suicidal behavior and childhood abuse and neglect: A meta-analysis. J Affect Disord 2017; 220: 147–55. [DOI] [PubMed] [Google Scholar]
- 26.Liu RT, Scopelliti KM, Pittman SK, Zamora AS. Childhood maltreatment and non-suicidal self-injury: a systematic review and meta-analysis. Lancet Psychiatry 2018; 5: 51–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Salokangas RKR, Luutonen S, Heinimaa M, From T, Hietala J. A study on the association of psychiatric diagnoses and childhood adversities with suicide risk. Nord J Psychiatry 2019; 73: 125–31. [DOI] [PubMed] [Google Scholar]
- 28.Bernegger A, Kienesberger K, Carlberg L, et al. Influence of Sex on Suicidal Phenotypes in Affective Disorder Patients with Traumatic Childhood Experiences. PloS One 2015; 10: e0137763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Devries KM, Mak JYT, Child JC, et al. Childhood sexual abuse and suicidal behavior: a meta-analysis. Pediatrics 2014; 133: e1331–1344. [DOI] [PubMed] [Google Scholar]
- 30.Nock MK, Hwang I, Sampson NA, Kessler RC. Mental disorders, comorbidity and suicidal behavior: results from the National Comorbidity Survey Replication. Mol Psychiatry 2010; 15: 868–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Kessler RC. The World Health Organization International Consortium in Psychiatric Epidemiology (ICPE): initial work and future directions -- the NAPE Lecture 1998. Nordic Association for Psychiatric Epidemiology. Acta PsychiatrScand 1999; 99: 2–9. [DOI] [PubMed] [Google Scholar]
- 32.Johnson TP, Wislar JS. Response rates and nonresponse errors in surveys. JAMA 2012; 307: 1805–6. [DOI] [PubMed] [Google Scholar]
- 33.Alonso J, Angermeyer MC, Bernert S, et al. Sampling and methods of the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta PsychiatrScand 2004; 109: 8–20. [DOI] [PubMed] [Google Scholar]
- 34.Brewin CR, Andrews B, Gotlib IH. Psychopathology and early experience: a reappraisal of retrospective reports. Psychol Bull 1993; 113: 82–98. [DOI] [PubMed] [Google Scholar]
- 35.Hardt J, Rutter M. Validity of adult retrospective reports of adverse childhood experiences: review of the evidence. J Child Psychol Psychiatry 2004; 45: 260–73. [DOI] [PubMed] [Google Scholar]
- 36.Kendler KS, Gallagher TJ, Abelson JM, Kessler RC. Lifetime prevalence, demographic risk factors, and diagnostic validity of nonaffective psychosis as assessed in a US community sample. The National Comorbidity Survey. ArchGenPsychiatry 1996; 53: 1022–31. [DOI] [PubMed] [Google Scholar]
- 37.Spengler PA, Wittchen HU. Procedural validity of standardized symptom questions for the assessment of psychotic symptoms--a comparison of the DIS with two clinical methods. ComprPsychiatry 1988; 29: 309–22. [DOI] [PubMed] [Google Scholar]
- 38.O’Connor RC, Nock MK. The psychology of suicidal behaviour. Lancet Psychiatry 2014; 1: 73–85. [DOI] [PubMed] [Google Scholar]
- 39.Glenn CR, Cha CB, Kleiman EM, Nock MK. Understanding Suicide Risk within the Research Domain Criteria (RDoC) Framework: Insights, Challenges, and Future Research Considerations. Clin Psychol Sci J Assoc Psychol Sci 2017; 5: 568–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Sáiz PA, Bobes J. Suicide prevention in Spain: an uncovered clinical need. Rev Psiquiatr Salud Ment 2014; 7: 1–4. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplement table 1. Bivariate analyses of risk factors (sociodemographic factors and mental disorders) for first onset of suicidal behaviors in Murcia (Spain).
Supplement table 3. Lifetime Mental Disorders as risk factors for first onset of suicidal behaviors in Murcia (Spain) in multivariate logistic regression models.
Supplement table 2. Bivariate analyses of Childhood adversities for first onset of suicidal behaviors in Murcia (Spain).
Data Availability Statement
Public access to the diagnostic instrument, including diagnostic algorithms, should be submitted to: http://www.hcp.med.harvard.edu/wmh. However, there are limitations on the availability of raw data due to ethical restrictions related to the signed consent agreements and to the signed agreement with the WHO World Mental Health Survey Initiative to limit comparative analyses to those carried out within the consortium. Requestors wishing to access a de-identified minimal dataset necessary for the purpose of monitoring our published analyses only, can apply to the principal investigator of the PEGASUS-Murcia Project, Fernando Navarro-Mateu (email: Fernando.navarro@carm.es).