Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2007 Sep 6.
Published in final edited form as: Psychiatry Res. 2007 Mar 13;152(1):11–20. doi: 10.1016/j.psychres.2006.09.003

Cholesterol and suicide attempts: A prospective study of depressed inpatients

Jess G Fiedorowicz a,*, William H Coryell a
PMCID: PMC1965495  NIHMSID: NIHMS27776  PMID: 17360043

Abstract

Low cholesterol levels have commonly been associated with various suicide measures. We sought to examine suicide attempts in a prospective sample of depressed patients that on prior analysis demonstrated an association between low cholesterol and subsequent suicide completions. Seventy-four inpatients with Research Diagnostic Criteria unipolar major depression, bipolar depression or schizoaffective depression entered a prospective follow-up study from 1978 - 1981. Kaplan-Meier survival analysis and Cox regression were utilized to elucidate the relationship between cholesterol levels and subsequent severe suicide attempts as well as all suicide attempts regardless of severity. Attempts preceding index hospitalization and other lifetime attempts were evaluated cross-sectionally. Low serum cholesterol levels did not predict subsequent suicide attempts. Contrary to our hypothesis, the high cholesterol group was associated with increased risk of suicide attempts on survival analysis in those less than median age. Nonetheless, in cross-sectional analysis, the low cholesterol group had more suicide attempts preceding index hospitalization and more remote lifetime attempts. The results from this prospective dataset do not support an association between low cholesterol and subsequent suicide attempts despite replicating the retrospective findings of previous case-control and cross-sectional studies.

Keywords: adult, major depression, bipolar depression, suicide, prospective cohort study, risk factors

1. Introduction

Suicide was the fifth leading cause of death based on years of potential life lost before age 65 in the United States from 1999-2002, the most recent years for which data is publicly available (CDC, 2005). Depressive disorders are the diagnoses uncovered most often in psychological autopsy studies (Barraclough et al., 1974; Rich et al., 1986; Henriksson et al., 1993; Isometsa et al., 1994; Conwell et al., 1996) and extensive efforts have been employed to elucidate risk factors for suicide among depressed patients to aid clinical decisions regarding appropriate level of care (Goldstein et al., 1991; APA, 2003; Coryell and Young, 2005).

In the early 1990's, analysis of mortality rates in primary prevention clinical trials revealed associations between low cholesterol and unnatural death (Muldoon et al., 1990; Lindberg et al., 1992; Neaton et al., 1992). Subsequent analyses of data from non-interventional community studies replicated and clarified this relationship. Of large (N > 5000) studies that described suicide as a secondary outcome, five found an association between low cholesterol levels and subsequent suicides (Lindberg et al.,, 1992; Neaton et al.,, 1992; Zureik et al., 1996; Partonen et al., 1999; Ellison and Morrison, 2001). Two, however, found no difference (Smith et al., 1992; Tamosiunas et al., 2005) and two demonstrated an inverse association (Iribarren et al., 1995; Tanskanen et al., 2000).

Case-control and cross-sectional studies have demonstrated lower cholesterol levels in patients admitted following suicide attempts compared to those admitted without an attempt (Takei et al., 1994; Kunugi et al., 1997; Papassotiropoulos et al., 1999; Garland et al., 2000; Sarchiapone et al., 2001; Guillem et al., 2002; Kim et al., 2002; Lee and Kim, 2003). Comparisons of suicide attempters to well controls have yielded similar results (Gallerani et al., 1995; Alvarez et al., 1999; Sarchiapone et al., 2000; Atmaca et al., 2002b; Kim et al.,, 2002; Tripodianakis et al., 2002; Lee and Kim, 2003). Some have found the association between low cholesterol and suicide attempts to be strongest among those with violent suicide attempts or with a history of such attempts (Alvarez et al., 2000; Bocchetta et al., 2001; Atmaca et al., 2003b; Vevera et al., 2003; Kim and Myint, 2004; Marcinko et al., 2005). Lower grey matter cholesterol has been demonstrated with violent suicide completers on autopsy (Lalovic et al., 2006) and it has even been suggested that a causal relationship exists between low cholesterol and violence in general (Golomb, 1998).

Nonetheless, many case control studies have found no relationship between cholesterol levels and recent or remote suicide attempts (Fritze et al., 1992; Engstrom et al., 1995; Seefried and Gumpel, 1997; Almeida-Montes et al., 2000; Huang and Wu, 2000; Huang, 2001; Roy et al., 2001; Tsai et al., 2002; Deisenhammer et al., 2004). One recent, large cross-sectional study demonstrated a relationship to elevated cholesterol and suicide attempts in the past year (Brunner et al., 2006) while another demonstrated an association between low high-density lipoprotein cholesterol and increased lifetime suicide attempts in women only (Zhang et al., 2005).

Case-control studies have also associated low cholesterol with other measures of suicidality including suicidal thoughts, suicidal intent, suicidal plan, or previous history of suicide attempts (Modai et al., 1994; Sullivan et al., 1994; Golier et al., 1995; Papassotiropoulos et al.,, 1999). An association between low cholesterol and suicidality is not limited to individuals with major depression but has also been described in patients with borderline personality disorder (Atmaca et al., 2002a); anorexia nervosa (Favaro et al., 2004); panic disorder (Ozer et al., 2004); and schizophrenia (Atmaca et al., 2003a).

Given the complex interplay of risk factors for suicide and the relative infrequency of suicide completions, the relationship between suicidality, suicide attempts, and suicide completions is not straightforward. Other variables associated with suicide attempts including the severity and timing of attempt (i.e. current episode versus remote) further complicate the comparison of studies.

Previous studies have not evaluated severity and timing broadly and have not integrated prospective and retrospective data. To our knowledge, there have been no prospective studies of cholesterol levels as a predictor of severe suicide attempts in a depressed cohort. The current report utilizes prospective data from the Iowa site of the National Institute of Mental Health (NIHM) Collaborative Study of Depression to extend to severe suicide attempts the results of an earlier analysis of the same population that demonstrated a relationship between low cholesterol and suicide completions (Coryell and Schlesser, in press).

2. Methods

2.1. Subjects

English-speaking, Caucasian adults who sought inpatient and outpatient treatment and met Research Diagnostic Criteria (RDC) (Spitzer et al., 1978) for major depressive disorder, mania, or schizoaffective disorder were recruited for the NIMH Collaborative Study of Depression (CSD) between the years 1978 and 1981 and were followed prospectively thereafter. The current analysis employs the sample used in an earlier report on completed suicide (Coryell and Schlesser, 2001). Four of 78 participants from this original sample did not have baseline cholesterol data and our current analysis was subsequently restricted to the 74 CSD participants at the Iowa site who had both a fasting cholesterol determination and a dexamethasone suppression test during their index admission. Data regarding use of cholesterol-lowering treatments during the time of intake is not available and likely negligible given an estimated 1980-1982 prevalence of use of 0.6% in women and 1% in men based on the Minnesota Heart Survey (Arnett et al., 2002).

2.2. Procedures

Initial diagnoses were based on the complete Schedule for Affective Disorders and Schizophrenia (SADS) (Endicott and Spitzer, 1978) and structured follow-up interviews using the Longitudinal Interval Follow-Up Evaluation (LIFE) (Keller et al., 1992). These occurred at six-month intervals for the first five years and yearly thereafter. The LIFE included characterizations of suicide attempts identical to the corresponding items in the SADS.

A serious attempt was operationally defined by the investigators prior to data analysis as a SADS suicidal intent or medical lethality score of ≥ 4. Table 1 details these specific SADS items of interest. Secondary analyses of the data ascertained the relationship between cholesterol levels and any subsequent suicide attempts, suicide attempts preceding index hospitalization, and other lifetime suicide attempts. Since this study was not anticipated at the time of data collection, raters were not formally blinded to cholesterol levels; however, raters did not assess cholesterol data.

Table 1. SADS Items Related to Suicide Attempts.

The specific wording of the relevant SADS items for severity of suicide attempts. The a priori cutoff for severe attempts is delineated.

Suicidal Intent:
Seriousness of suicidal intent to kill self
as judged by overall circumstances
including likelihood of being rescued,
precautions against discovery, action to
gain help during or after attempt, degree
of planning and the apparent purpose of
attempt, degree of planning and the
apparent purpose of attempt
(manipulative versus killing self)

  • 0 No information

  • 1 Obviously no intent, purely manipulative gestures

  • 2 Not sure or only minimal intent

  • 3 Definite but very ambivalent

--------------------(cutoff)--------------------

  • 4 Serious

  • 5 Very Serious

  • 6 Extreme (every expectation of death)
Medical Lethality:
Actual medical threat to life or
physical condition following the most
serious suicide gesture, taking into
account the method (gunshot wound
more serious than knife wound),
impaired consciousness at or during
time of rescue, seriousness of lesion,
toxicity of ingested materials,
reversibility (amount of time expected
for complete recovery), and amount of
treatment required.

  • 0 No information

  • 1 No danger, e.g., no effects, held pills in hand

  • 2 Minimal, e.g., scratch on wrist

  • 3 Mild, e.g., took 10 aspirins, mild gastritis

--------------------(cutoff)--------------------

  • 4 Moderate, e.g., took 10 Seconals, had brief unconsciousness

  • 5 Severe, e.g., cuts throat

  • 6 Extreme, e.g., respiratory arrest or prolonged coma
Open in a new tab

2.3. Data Analysis

Previous studies of cholesterol and subsequent suicide have divided cholesterol levels into two to five groups. To facilitate comparisons to a previous analysis of suicide completions in the same sample (Coryell and Schlesser, in press), cholesterol values were divided into two groups with the lower range consisting of values less than or equal to 190 mg/dl, representing approximately the lower tertile and selected from a receiver operator characteristic (ROC) curve.

Additionally, an ROC curve constructed to determine an optimal cholesterol cutoff point related to suicide attempts failed to select a superior cutoff. To facilitate clarity of communication, the groups representing the lower tertile and upper tertiles are referred to as the “low” and “high” cholesterol groups respectively.

Severe suicide attempts were operationally defined a priori as attempts that reached a score of ≥ 4 for either intent or medical lethality based on the SADS. A further post hoc analysis looked at all recorded suicide attempts (i.e. SADS ≥ 1). Higher and lower cholesterol groups were compared with nonparametric Kaplan-Meier survival analysis, which included censored data resulting from loss to follow-up. Median survival could not be reported in this sample as a median cumulative survival of 0.5 was not reached during follow-up. A Cox Regression Analysis was further performed to adjust for potential confounds.

Additionally, a cross-sectional analysis was performed to look at immediately preceding suicide attempts and other lifetime suicide attempts for comparison to previous case-control and cross-sectional studies. The cross-sectional data was analyzed using an independent samples t-test with cholesterol as a continuous variable and chi-square with cholesterol as a dichotomous variable. All statistical tests were two-tailed.

3. Results

Cholesterol concentrations increased significantly with age (r = 0.5, P < 0.0001) and age was subsequently used as a covariate in regression analyses. A comparison of clinical characteristics between the high and low cholesterol groups demonstrated a significant age difference between groups (t = −3.8, df = 72, P = 0.0003). The high cholesterol group also had more individuals with hypertension and hypothyroidism. No further significant differences were observed between groups.

3.1. Cholesterol and Future Severe Suicide Attempts

The mean (SD) total cholesterol concentration for the entire sample of seventy-four patients was 210.54 (44.67) mg/dl with a range of 109 to 360, consistent with published populations norms of the region and time period (Arnett et al.,, 2002). The 18 patients with serious suicide attempts had a mean (SD) cholesterol level of 213.4 (36.8) mg/dl and the remaining 56 patients had a mean (SD) cholesterol level of 209.6 (47.2) mg/dl (t = −0.3, df = 72, P = 0.75). A Kaplan-Meier survival analysis (Figure 1) did not support the predicted relationship between lower cholesterol levels and the likelihood of later suicide attempts. The cumulative survival estimation for those with low cholesterol concentrations was 83.3 % compared to 65.7% in the high cholesterol group. These survival rates precluded calculation of median survival time. The mean (SD) survival time for the high cholesterol group was 574 (45) weeks (95% confidence interval 485 – 663 weeks). The mean (SD) survival time for the low cholesterol group was 690 (50) weeks (95% confidence interval 593 – 787 weeks). The mean (SD) number of serious attempts per decade of follow-up were 0.73 (1.46) and 0.39 (1.13) for the high and low cholesterol groups respectively (t = −1.0, df = 72.0, P = 0.32; age-adjusted F = 4.4, P = 0.04).

Figure 1.

Figure 1

Open in a new tab

Kaplan-Meier Survival Analysis comparing cumulative survival based on occurrence of severe suicide attempt for high and low total cholesterol groups over 15 years (780 weeks) of follow-up.

In survival analysis of time until the first attempt adjusted for age, the high cholesterol group experienced more severe suicide attempts. (Hazard Ratio 7.82, 95% C.I. 2.00 – 30.77, P = 0.003). A significant interaction between age and cholesterol level was also observed (P = 0.001). The presence of prior severe attempts were nearly correlated with future severe attempts (r = 0.2, P = 0.057). With prior severe attempts, age, and the age-cholesterol interaction included as covariates, a significant age by cholesterol group interaction between the high and low cholesterol groups persisted (P = 0.001). When stratified by the median age of 32 years, the high cholesterol group remained at greater risk of suicide attempt in those below median age (Hazard Ratio 19.38, 95% C.I. 0.2.49 – 149.58, P = 0.005) while no significant differences were noted in those above median age (Hazard Ratio 0.16, 95% C.I. 0.02 – 1.14, P = 0.007). Admission body mass index, gender, hypertension, hyperthyroidism, marital status and psychiatric diagnosis did not result in any substantial changes in the hazard ratios when added as covariates.

3.2. Cholesterol and All Future Suicide Attempts

The twenty-three patients with any future suicide attempt had a mean (SD) cholesterol level of 213.6 (33.4) mg/dl and the remaining fifty-one patients had a mean (SD) cholesterol level of 209.2 (49.2) mg/dl (t = −0.4, df = 72, P = 0.70). A Kaplan-Meier survival analysis (Figure 2) was performed and demonstrated increased suicide attempts in the high cholesterol group, contrary to the prediction of a greater frequency of severe suicide attempts in the low cholesterol group. The cumulative survival estimation for those with low cholesterol concentrations was 84.4% compared to 54.1% in the high cholesterol group. These survival rates precluded calculation of median survival time. The mean (SD) survival time for the high cholesterol group was 511 (47) weeks (95% confidence interval 418 – 603 weeks). The mean (SD) survival time for the low cholesterol group was 675 (56) weeks (95% confidence interval 565 – 785 weeks). There were significantly more attempts per decade of follow-up in the high cholesterol group number with a mean (SD) of 1.67 (3.56) attempts per decade compared to 0.42 (1.19) attempts per decade in the low cholesterol group (t = −2.3, df = 68.4, p = 0.028; age-adjusted F = 5.8, P = 0.019).

Figure 2.

Figure 2

Open in a new tab

Kaplan-Meier Survival Analysis comparing cumulative survival based on occurrence of any suicide attempt for high and low total cholesterol groups over 15 years (780 weeks) of follow-up.

In survival analysis of time until the first attempt adjusted for age, the high cholesterol group experienced more suicide attempts. (Hazard Ratio 6.85, 95% C.I. 1.85 – 25.38, P = 0.004). A significant interaction between age and cholesterol was again observed (Hazard Ratio 0.84, 95% C.I. 0.73 – 0.97, P = 0.017). Prior attempts did not correlate with future severe attempts (r = 0.04, P = 0.76). When stratified by the median age of 32 years, the high cholesterol group remained at greater risk of suicide attempt in those below median age (Hazard Ratio 18.51, 95% C.I. 2.38 – 144.14, P = 0.005) while no significant differences were noted in those above median age (Hazard Ratio 0.52, 95% C.I. 0.11 – 2.52, P = 0.42). Admission body mass index, gender, hypertension, hyperthyroidism, marital status and psychiatric diagnosis did not result in any substantial changes in the hazard ratios when added as covariates.

3.3. Cholesterol and Recent Suicide Attempts

A cross-sectional study was performed on this cohort of 74 subjects of which 20 had a severe suicide attempt and 27 had any suicide attempt during the index depressive episode prior to admission or within the past year if the index depressive episode exceeded one year in duration. The 20 patients with severe preceding suicide attempts had a mean (SD) cholesterol level of 187.8 (46.3) mg/dl and the 54 without preceding severe suicide attempts had a mean (SD) cholesterol level of 219.0 (41.4) mg/dl (t = 2.8, df = 72, P = 0.007), which retains significance when age is included as a covariate (F = 4.6, P = 0.035). The 27 patients with any preceding suicide attempt had a mean (SD) cholesterol level of 191.9 (42.4) mg/dl and the 47 patients without any preceding suicide attempt had a mean (SD) cholesterol level of 221.2 (42.8) mg/dl (t = 2.8, df = 72, P = 0.006), which retains significance when age is included as a covariate (F = 5.2, P = 0.026).

Ten of 23 (43.8%) patients in the low cholesterol group and 10 of 51 (19.6%) patients in the high cholesterol group had a severe suicide attempt as defined by our aforementioned criteria preceding index admission and during the index depressive episode, resulting in a statistically significant difference between groups (χ2 = 4.6, df = 1, P = 0.032), which retains significance when age is included as a covariate (Hazard Ratio 8.77, 95% C.I. 1.80 – 41.67, P = 0.007). Data regarding the total number of severe suicide attempts during index depressive episode per subject could not be extrapolated from the data set.

Thirteen of 23 (56.5%) patients in the low cholesterol group and 14 of 51 (27.5%) of the patients in the high cholesterol group had any suicide attempt prior to admission during the same depressive episode, resulting in a statistically significant difference between groups (χ2 = 5.8, df = 1, P = 0.016), which retains significance when age is included as a covariate (Hazard Ratio 7.25, 95% C.I. 1.69 – 31.25, P = 0.008). There was also a significant difference in the number of total attempts between groups (χ2 = 10.4, df = 4, P = 0.035; age-adjusted F = 0.58, P = 0.019).

3.4. Cholesterol and Remote Lifetime Suicide Attempts

A cross-sectional study was again performed on this cohort of 74 subjects of which 12 had a serious suicide attempt and 24 had any suicide attempt in their lifetime prior to the index depressive episode or prior to the past year if the index depressive episode exceeded one year in duration. The 12 patients with severe remote lifetime suicide attempts had a mean (SD) cholesterol level of 193.4 (52.7) mg/dl and the 62 without a history of such attempts had a mean (SD) cholesterol level of 213.9 (42.6) mg/dl (t = 1.5, df = 72, P = 0.15; age-adjusted F = 0.26, P = 0.61). The 24 patients with any remote lifetime suicide attempt had a mean (SD) cholesterol level of 200.4 (43.7) mg/dl and the 50 patients without such history had a mean (SD) cholesterol level of 215.4 (44.7) mg/dl (t = 1.4, df = 72, P = 0.18; age-adjusted F = 0.23, P = 0.64).

Six of 23 (26.1%) patients in the low cholesterol group and 6 of 51 (11.8%) patients in the high cholesterol group had a remote lifetime severe suicide attempt as defined by our aforementioned criteria (χ2 = 2.4, df = 1, P = 0.12; age-adjusted P = 0.65). Data regarding the total number of remote lifetime severe suicide attempts per subject could not be extrapolated from the data set.

Ten of 23 (43.5%) patients in the low cholesterol group and 14 of 51 (27.5%) of the patients in the high cholesterol group had any remote suicide attempt (χ2 = 1.9, df = 1, P = 0.17; age-adjusted P = 0.62). The low cholesterol group had a significantly greater number of lifetime suicide attempts prior to index depressive episode (χ2 = 12.1, df = 5, P = 0.034; age-adjusted F = 4.6, P = 0.035).

4. Discussion

In this prospective sample of depressed patients, an association was not found between severe suicide attempts and low cholesterol levels. This finding is in contrast with a previous analysis of the same data set demonstrating an association between low cholesterol levels and suicide completions (Coryell and Schlesser, in press). The current analysis not only fails to support the initial hypothesized association between low baseline cholesterol levels and subsequent severe suicide attempts, it suggests the contrary, a finding which persists when extended to include all suicide attempts. Exploration of a cholesterol group by age interaction further reveals that the high cholesterol group has more suicide attempts in only those below median age.

Some studies have suggested that low cholesterol is a consequence of depression, perhaps secondary to diminished appetite or weight loss, and that this accounts for the observed association (Law et al., 1994). However, later studies controlled for weight (Golier et al.,, 1995), serum total protein (Kunugi et al.,, 1997), or excluded significant weight loss in the month prior to admission (Garland et al.,, 2000) and nevertheless linked low cholesterol concentrations to suicidal behavior. Lower serotonergic function may mediate this link (Golomb et al., 2002) and depletion of membrane cholesterol has been demonstrated to adversely effect serotonin transporter activity (Scanlon et al., 2001). Additionally, cholesterol may influence illness and behavior via its important role as a component of the myelin sheath, a precursor for steroid hormone synthesis, and a constituent of cell membranes with subsequent roles in transmembrane exchange, enzyme function, and neurotransmitter receptor expression (Golomb et al., 2004).

The present study is limited by its sample size, limited observations of suicide attempts, and lack of cholesterol data during follow-up. Additionally, the subjects studied were all recruited as inpatients. While this may limits the generalizability, the sample arguably represents a population with a higher suicide risk and subsequently greater applied predictive value for suicide prediction measures (Qin et al., 2003; Sinclair et al., 2005). Furthermore, while data collection regarding time to first attempt is generally accurate, our data related to the total number of attempts may underestimate total attempts. During some periods of observation, protocols recommended reporting only the most severe subsequent attempts for each specific follow-up period. Age presented a significant confound for comparison between high and low cholesterol groups. The positive correlation between cholesterol levels and age is consistent with other studies (Fritze et al.,, 1992; Garland et al.,, 2000; Bocchetta et al.,, 2001) and the natural history of hyperlipidemia with age as a non-modifiable risk factor (Nestruck and Davignon, 1986). Analyses were subsequently controlled for age.

The use of a prospective cohort of depressed subjects in the current study poses several advantages. The predictive value of a screening measure is largely influenced by prevalence. Given the higher rate of suicide in patients with mood disorders compared to the general population, this study concerns precisely the population to which predictors of suicide are most often applied. Furthermore, the use of a prospective sample, with cholesterol levels determined prior to any suicide attempts, allows for a temporal relationship to be established with cholesterol levels preceding attempts. Cross-sectional studies are hindered by measuring cholesterol levels after suicide attempts. The results of cross-sectional studies, therefore, cannot be applied as readily to a prediction model and cannot differentiate low cholesterol levels which precede the outcome measure from those which are the result of the outcome or underlying condition associated with the outcome.

These unexpected results of our current analysis are difficult to reconcile with the aforementioned findings of previous studies, particularly those of the case-control and cross-sectional studies. This motivated the cross-sectional analysis, which ultimately replicated previously published findings by demonstrating lower cholesterol levels among those admitted following suicide attempt. Within this same prospective data set the following associations have been found: 1) low cholesterol levels with subsequent suicide completions, 2) high cholesterol levels with subsequent suicide attempts, including severe suicide attempts in those less than median age, and 3) prior suicide attempts with low cholesterol levels. While these results may simply be an artifact of dichotomization, the contrasting findings compel further explanation. A variety of potential explanations may be posed for these seemingly discordant findings. The above analyses include the use of different suicide outcome measures in association 1 (suicide completions) versus associations 2 and 3 (suicide attempts). These suicide outcome measures and suicidality markers are not directly interchangeable. The above analyses also subsume different temporal relationships between the measurement of cholesterol and assessed outcome in association 3 (retrospective) versus associations 1 and 2 (prospective). Thus cross-sectional findings may not apply prospectively, suggesting that low cholesterol may potentially result from or be an epiphenomenon of suicide attempts or, more likely, the underlying pathology motivating suicide attempt. This is further supported by the greater magnitude of association between low cholesterol and recent attempts versus remote attempts. Admittedly, this explanation does not completely address the association between low cholesterol and subsequent suicide completions. It is furthermore not clear why association 2 appears restricted to the younger half of the sample.

These findings illustrate the difficulty inherent to the study of the spectrum of suicidality and suicidal behaviors. While intuitively suicidal ideation may serve a surrogate for suicide attempts and subsequently completed suicides, this association is a weak one (Young and Coryell, 2005). A 21 year follow-up study of 785 patients with affective disorders showed that while completed suicide was more likely in those with a history of suicide attempts than in those without such history (8.5% versus 3.5%), the vast majority of completions were committed by those without a history of prior attempts. Furthermore, 97 of 106 attempters did not go on to complete suicide (Coryell and Young, 2005).

The current findings may, in part, explain some of the controversy in the literature regarding the relationship of cholesterol and other biological markers to suicide. The disparity in the relationship between cholesterol and suicide completions versus suicide attempts is not without precedent. A parallel may be drawn to gender differences in suicidal behaviors with more attempts by women and more completions by men (Kuo et al., 2001; APA, 2003). Our findings caution the at times reflexive substitution of suicidal thoughts or suicide attempts as a surrogate for risk of completed suicide. A contemporaneous example of this may include the debate over the risk of suicide with antidepressants in children and adolescents (Brent, 2004; Ryan, 2005). Nevertheless, this caution applies to any inductive argument linking an association of a variable with suicidality to suicide completions solely based on the association between suicidality and suicide completions.

Table 2. Demographic and Clinical Characteristics at Baseline by Total Cholesterol Level.

Demographic and clinical characteristics of high and low cholesterol groups. The high cholesterol group contains cholesterol levels > 190 mg/dL. Independent samples t-tests and chi-square tests were utilized to assess differences between groups. No statistically significant differences were found other than age at baseline, prevalence of hypertension, and prevalence of hypothyroidism.

Characteristic High Cholesterol
(N=51)		 Low Cholesterol
(N=23)
Mean SD Mean SD
Length of follow-up (weeks) 538 278 542 299
Age at baseline (years)a 39.8 14.8 27 8.5
N % N %
Female Gender 34 66.7% 14 60.9%
  Intake Psychiatric Diagnosis:b
Major depressive disorder 37 72.5% 15 65.2%
Bipolar type II, depressed 6 11.8% 3 13.0%
Bipolar type I, depressed 5 9.8% 5 21.7%
Schizoaffective disorder, depressed 3 5.9% 0 0.0%
History of Alcohol Abuse 13 25.5% 9 39.1%
History of Illicit Drug Abuse 2 3.9% 1 4.3%
Anorexia Nervosa 1 2.0% 1 4.3%
  Medical Co-morbidities at Intake:
Diabetes mellitus 1 2.0% 0 0.0%
Hypertensionc 10 19.6% 0 0.0%
Hypothyroidismc 8 15.7% 1 4.3%
History of Myocardial Infarction 1 2.0% 0 0.0%
Open in a new tab
a

Statistically significant p < 0.001.

b

Diagnoses made with the Schedule for Affective Disorders and Schizophrenia.

c

Statistically significant p < 0.05.

Acknowledgements

We would like to thank Dr. Joseph E. Cavanaugh from the University of Iowa, College of Public Health, Department of Biostatistics for lending his statistical expertise and Dr. Valerie Forman Hoffman from the University of Iowa, Iowa Scholars in Clinical Investigation Program for her epidemiological expertise in review of the manuscript. We would also like to thank Carol A. Moss and Barbara M. Robb for their assistance with data management. This study was funded by NIMH grant MH 25416.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Almeida-Montes LG, Valles-Sanchez V, Moreno-Aguilar J, Chavez-Balderas RA, Garcia-Marin JA, Cortes Sotres JF, Hheinze-Martin G. Relation of serum cholesterol, lipid, serotonin and tryptophan levels to severity of depression and to suicide attempts. Journal of Psychiatry & Neuroscience. 2000;25:371–377. [PMC free article] [PubMed] [Google Scholar]
  2. Alvarez JC, Cremniter D, Gluck N, Quintin P, Leboyer M, Berlin I, Therond P, Spreux-Varoquaux O. Low serum cholesterol in violent but not in non-violent suicide attempters. Psychiatry Research. 2000;95:103–108. doi: 10.1016/s0165-1781(00)00171-2. [DOI] [PubMed] [Google Scholar]
  3. Alvarez JC, Cremniter D, Lesieur P, Gregoire A, Gilton A, Macquin-Mavier I, Jarreau C, Spreux-Varoquaux O. Low blood cholesterol and low platelet serotonin levels in violent suicide attempters. Biological Psychiatry. 1999;45:1066–1069. doi: 10.1016/s0006-3223(98)00160-7. [DOI] [PubMed] [Google Scholar]
  4. APA Practice guideline for the assessment and treatment of patients with suicidal behaviors. American Journal of Psychiatry. 2003;160:1–60. [PubMed] [Google Scholar]
  5. Arnett DK, McGovern PG, Jacobs DR, Jr., Shahar E, Duval S, Blackburn H, Luepker RV. Fifteen-year trends in cardiovascular risk factors (1980-1982 through 1995-1997): the Minnesota Heart Survey. American Journal of Epidemiology. 2002;156:929–935. doi: 10.1093/aje/kwf133. [DOI] [PubMed] [Google Scholar]
  6. Atmaca M, Kuloglu M, Tezcan E, Gecici O, Ustundag B. Serum cholesterol and leptin levels in patients with borderline personality disorder. Neuropsychobiology. 2002a;45:167–171. doi: 10.1159/000063665. [DOI] [PubMed] [Google Scholar]
  7. Atmaca M, Kuloglu M, Tezcan E, Ustundag B. Serum leptin and cholesterol levels in schizophrenic patients with and without suicide attempts. Acta Psychiatrica Scandinavica. 2003a;108:208–214. doi: 10.1034/j.1600-0447.2003.00145.x. [DOI] [PubMed] [Google Scholar]
  8. Atmaca M, Kuloglu M, Tezcan E, Ustundag B. Serum leptin and cholesterol levels in schizophrenic patients with and without suicide attempts. Acta Psychiatrica Scandinavica. 2003b;108:208–214. doi: 10.1034/j.1600-0447.2003.00145.x. [DOI] [PubMed] [Google Scholar]
  9. Atmaca M, Kuloglu M, Tezcan E, Ustundag B, Gecici O, Firidin B. Serum leptin and cholesterol values in suicide attempters. Neuropsychobiology. 2002b;45:124–127. doi: 10.1159/000054950. [DOI] [PubMed] [Google Scholar]
  10. Barraclough B, Bunch J, Nelson B, Sainsbury P. A hundred cases of suicide: clinical aspects. British Journal of Psychiatry. 1974;125:355–373. doi: 10.1192/bjp.125.4.355. [DOI] [PubMed] [Google Scholar]
  11. Bocchetta A, Chillotti C, Carboni G, Oi A, Ponti M, Del Zompo M. Association of personal and familial suicide risk with low serum cholesterol concentration in male lithium patients. Acta Psychiatrica Scandinavica. 2001;104:37–41. doi: 10.1034/j.1600-0447.2001.00374.x. [DOI] [PubMed] [Google Scholar]
  12. Brent DA. Antidepressants and pediatric depression--the risk of doing nothing. New England Journal of Medicine. 2004;351:1598–1601. doi: 10.1056/NEJMp048228. [DOI] [PubMed] [Google Scholar]
  13. Brunner J, Bronisch T, Pfister H, Jacobi F, Hofler M, Wittchen HU. High cholesterol, triglycerides, and body-mass index in suicide attempters. Archives of Suicide Research. 2006;10:1–9. doi: 10.1080/13811110500318083. [DOI] [PubMed] [Google Scholar]
  14. CDC Years of Potential Life Lost (YPLL) Before Age 65, 1999 - 2002 United States, WISQARS Years of Potential Life Lost (YPLL) Reports, 1999 - 2002. 2005 Retrieved on 10/21/2005 from http://webapp.cdc.gov/sasweb/ncipc/ypll10.html.
  15. Conwell Y, Duberstein PR, Cox C, Herrmann JH, Forbes NT, Caine ED. Relationships of age and axis I diagnoses in victims of completed suicide: a psychological autopsy study. American Journal of Psychiatry. 1996;153:1001–1008. doi: 10.1176/ajp.153.8.1001. [DOI] [PubMed] [Google Scholar]
  16. Coryell W, Schlesser M. The dexamethasone suppression test and suicide prediction. American Journal of Psychiatry. 2001;158:748–753. doi: 10.1176/appi.ajp.158.5.748. [DOI] [PubMed] [Google Scholar]
  17. Coryell W, Schlesser M. Combined Biological Tests for Suicide Prediction. Psychiatry Research. doi: 10.1016/j.psychres.2006.01.021. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Coryell W, Young EA. Clinical predictors of suicide in primary major depressive disorder. Journal of Clinical Psychiatry. 2005;66:412–417. doi: 10.4088/jcp.v66n0401. [DOI] [PubMed] [Google Scholar]
  19. Deisenhammer EA, Kramer-Reinstadler K, Liensberger D, Kemmler G, Hinterhuber H, Fleischhacker WW. No evidence for an association between serum cholesterol and the course of depression and suicidality. Psychiatry Research. 2004;121:253–261. doi: 10.1016/j.psychres.2003.09.007. [DOI] [PubMed] [Google Scholar]
  20. Ellison LF, Morrison HI. Low serum cholesterol concentration and risk of suicide. Epidemiology. 2001;12:168–172. doi: 10.1097/00001648-200103000-00007. [DOI] [PubMed] [Google Scholar]
  21. Endicott J, Spitzer RL. A diagnostic interview: the schedule for affective disorders and schizophrenia. Archives of General Psychiatry. 1978;35:837–844. doi: 10.1001/archpsyc.1978.01770310043002. [DOI] [PubMed] [Google Scholar]
  22. Engstrom G, Alsen M, Regnell G, Traskman-Bendz L. Serum lipids in suicide attempters. Suicide and Life-Threatening Behavior. 1995;25:393–400. [PubMed] [Google Scholar]
  23. Favaro A, Caregaro L, Di Pascoli L, Brambilla F, Santonastaso P. Total serum cholesterol and suicidality in anorexia nervosa. Psychosomatic Medicine. 2004;66:548–552. doi: 10.1097/01.psy.0000127873.31062.80. [DOI] [PubMed] [Google Scholar]
  24. Fritze J, Schneider B, Lanczik M. Autoaggressive behaviour and cholesterol. Neuropsychobiology. 1992;26:180–181. doi: 10.1159/000118916. [DOI] [PubMed] [Google Scholar]
  25. Gallerani M, Manfredini R, Caracciolo S, Scapoli C, Molinari S, Fersini C. Serum cholesterol concentrations in parasuicide. British Medical Journal. 1995;310:1632–1636. doi: 10.1136/bmj.310.6995.1632. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Garland M, Hickey D, Corvin A, Golden J, Fitzpatrick P, Cunningham S, Walsh N. Total serum cholesterol in relation to psychological correlates in parasuicide. British Journal of Psychiatry. 2000;177:77–83. doi: 10.1192/bjp.177.1.77. [DOI] [PubMed] [Google Scholar]
  27. Goldstein RB, Black DW, Nasrallah A, Winokur G. The prediction of suicide. Sensitivity, specificity, and predictive value of a multivariate model applied to suicide among 1906 patients with affective disorders. Archives of General Psychiatry. 1991;48:418–422. doi: 10.1001/archpsyc.1991.01810290030004. [DOI] [PubMed] [Google Scholar]
  28. Golier JA, Marzuk PM, Leon AC, Weiner C, Tardiff K. Low serum cholesterol level and attempted suicide. American Journal of Psychiatry. 1995;152:419–423. doi: 10.1176/ajp.152.3.419. [DOI] [PubMed] [Google Scholar]
  29. Golomb BA. Cholesterol and violence: is there a connection? Annals of Internal Medicine. 1998;128:478–487. doi: 10.7326/0003-4819-128-6-199803150-00009. [DOI] [PubMed] [Google Scholar]
  30. Golomb BA, Criqui MH, White HL, Dimsdale JE. The UCSD Statin Study: a randomized controlled trial assessing the impact of statins on selected noncardiac outcomes. Controlled Clinical Trials. 2004;25:178–202. doi: 10.1016/j.cct.2003.08.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Golomb BA, Tenkanen L, Alikoski T, Niskanen T, Manninen V, Huttunen M, Mednick SA. Insulin sensitivity markers: predictors of accidents and suicides in Helsinki Heart Study screenees. Journal of Clinical Epidemiology. 2002;55:767–773. doi: 10.1016/s0895-4356(02)00407-9. [DOI] [PubMed] [Google Scholar]
  32. Guillem E, Pelissolo A, Notides C, Lepine JP. Relationship between attempted suicide, serum cholesterol level and novelty seeking in psychiatric in-patients. Psychiatry Research. 2002;112:83–88. doi: 10.1016/s0165-1781(02)00193-2. [DOI] [PubMed] [Google Scholar]
  33. Henriksson MM, Aro HM, Marttunen MJ, Heikkinen ME, Isometsa ET, Kuoppasalmi KI, Lonnqvist JK. Mental disorders and comorbidity in suicide. American Journal of Psychiatry. 1993;150:935–940. doi: 10.1176/ajp.150.6.935. [DOI] [PubMed] [Google Scholar]
  34. Huang T, Wu S. Serum cholesterol levels in paranoid and non-paranoid schizophrenia associated with physical violence or suicide attempts in Taiwanese. Psychiatry Research. 2000;96:175–178. doi: 10.1016/s0165-1781(00)00206-7. [DOI] [PubMed] [Google Scholar]
  35. Huang TL. Serum cholesterol levels in mood disorders associated with physical violence or suicide attempts in Taiwanese. Chang Gung Medical Journal. 2001;24:563–568. [PubMed] [Google Scholar]
  36. Iribarren C, Reed DM, Wergowske G, Burchfiel CM, Dwyer JH. Serum cholesterol level and mortality due to suicide and trauma in the Honolulu Heart Program. Archives of Internal Medicine. 1995;155:695–700. [PubMed] [Google Scholar]
  37. Isometsa ET, Henriksson MM, Aro HM, Lonnqvist JK. Suicide in bipolar disorder in Finland. American Journal of Psychiatry. 1994;151:1020–1024. doi: 10.1176/ajp.151.7.1020. [DOI] [PubMed] [Google Scholar]
  38. Keller MB, Lavori PW, Mueller TI, Endicott J, Coryell W, Hirschfeld RM, Shea T. Time to recovery, chronicity, and levels of psychopathology in major depression. A 5-year prospective follow-up of 431 subjects. Archives of General Psychiatry. 1992;49:809–816. doi: 10.1001/archpsyc.1992.01820100053010. [DOI] [PubMed] [Google Scholar]
  39. Kim YK, Lee HJ, Kim JY, Yoon DK, Choi SH, Lee MS. Low serum cholesterol is correlated to suicidality in a Korean sample. Acta Psychiatrica Scandinavica. 2002;105:141–148. doi: 10.1034/j.1600-0447.2002.10352.x. [DOI] [PubMed] [Google Scholar]
  40. Kim YK, Myint AM. Clinical application of low serum cholesterol as an indicator for suicide risk in major depression. Journal of Affective Disorders. 2004;81:161–166. doi: 10.1016/S0165-0327(03)00166-6. [DOI] [PubMed] [Google Scholar]
  41. Kunugi H, Takei N, Aoki H, Nanko S. Low serum cholesterol in suicide attempters. Biological Psychiatry. 1997;41:196–200. doi: 10.1016/S0006-3223(95)00672-9. [DOI] [PubMed] [Google Scholar]
  42. Kuo WH, Gallo JJ, Tien AY. Incidence of suicide ideation and attempts in adults: the 13-year follow-up of a community sample in Baltimore, Maryland. Psychological Medicine. 2001;31:1181–1191. doi: 10.1017/s0033291701004482. [DOI] [PubMed] [Google Scholar]
  43. Lalovic A, Levy E, Luheshi G, Canetti L, Grenier E, Sequeira A, Turecki G. Cholesterol content in brains of suicide completers. International Journal of Neuropsychopharmacology. 2006:1–8. doi: 10.1017/S1461145706006663. [DOI] [PubMed] [Google Scholar]
  44. Law MR, Thompson SG, Wald NJ. Assessing possible hazards of reducing serum cholesterol. British Medical Journal. 1994;308:373–379. doi: 10.1136/bmj.308.6925.373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Lee HJ, Kim YK. Serum lipid levels and suicide attempts. Acta Psychiatrica Scandavica. 2003;108:215–221. doi: 10.1034/j.1600-0447.2003.00115.x. [DOI] [PubMed] [Google Scholar]
  46. Lindberg G, Rastam L, Gullberg B, Eklund GA. Low serum cholesterol concentration and short term mortality from injuries in men and women. British Medical Journal. 1992;305:277–279. doi: 10.1136/bmj.305.6848.277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Marcinko D, Martinac M, Karlovic D, Filipcic I, Loncar C, Pivac N, Jakovljevic M. Are there differences in serum cholesterol and cortisol concentrations between violent and non-violent schizophrenic male suicide attempters? Collegium Anthropologicum. 2005;29:153–157. [PubMed] [Google Scholar]
  48. Modai I, Valevski A, Dror S, Weizman A. Serum cholesterol levels and suicidal tendencies in psychiatric inpatients. Journal of Clinical Psychiatry. 1994;55:252–254. [PubMed] [Google Scholar]
  49. Muldoon MF, Manuck SB, Matthews KA. Lowering cholesterol concentrations and mortality: a quantitative review of primary prevention trials. British Medical Journal. 1990;301:309–314. doi: 10.1136/bmj.301.6747.309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Neaton JD, Blackburn H, Jacobs D, Kuller L, Lee DJ, Sherwin R, Shih J, Stamler J, Wentworth D. Serum cholesterol level and mortality findings for men screened in the Multiple Risk Factor Intervention Trial. Multiple Risk Factor Intervention Trial Research Group. Archives of Internal Medicine. 1992;152:1490–1500. [PubMed] [Google Scholar]
  51. Nestruck AC, Davignon J. Risks for hyperlipidemia. Cardiology Clinics. 1986;4:47–56. [PubMed] [Google Scholar]
  52. Ozer OA, Kutanis R, Agargun MY, Besiroglu L, Bal AC, Selvi Y, Kara H. Serum lipid levels, suicidality, and panic disorder. Comprehensive Psychiatry. 2004;45:95–98. doi: 10.1016/j.comppsych.2003.12.004. [DOI] [PubMed] [Google Scholar]
  53. Papassotiropoulos A, Hawellek B, Frahnert C, Rao GS, Rao ML. The risk of acute suicidality in psychiatric inpatients increases with low plasma cholesterol. Pharmacopsychiatry. 1999;32:1–4. doi: 10.1055/s-2007-979181. [DOI] [PubMed] [Google Scholar]
  54. Partonen T, Haukka J, Virtamo J, Taylor PR, Lonnqvist J. Association of low serum total cholesterol with major depression and suicide. British Journal of Psychiatry. 1999;175:259–262. doi: 10.1192/bjp.175.3.259. [DOI] [PubMed] [Google Scholar]
  55. Qin P, Agerbo E, Mortensen PB. Suicide risk in relation to socioeconomic, demographic, psychiatric, and familial factors: a national register-based study of all suicides in Denmark, 1981-1997. American Journal of Psychiatry. 2003;160:765–772. doi: 10.1176/appi.ajp.160.4.765. [DOI] [PubMed] [Google Scholar]
  56. Rich CL, Young D, Fowler RC. San Diego suicide study. I. Young vs old subjects. Archives of General Psychiatry. 1986;43:577–582. doi: 10.1001/archpsyc.1986.01800060071009. [DOI] [PubMed] [Google Scholar]
  57. Roy A, Gonzalez B, Marcus A, Berman J. Serum cholesterol, suicidal behavior and impulsivity in cocaine-dependent patients. Psychiatry Research. 2001;101:243–247. doi: 10.1016/s0165-1781(01)00217-7. [DOI] [PubMed] [Google Scholar]
  58. Ryan ND. Treatment of depression in children and adolescents. Lancet. 2005;366:933–940. doi: 10.1016/S0140-6736(05)67321-7. [DOI] [PubMed] [Google Scholar]
  59. Sarchiapone M, Camardese G, Roy A, Della Casa S, Satta MA, Gonzalez B, Berman J, De Risio S. Cholesterol and serotonin indices in depressed and suicidal patients. Journal of Affective Disorders. 2001;62:217–219. doi: 10.1016/s0165-0327(99)00200-1. [DOI] [PubMed] [Google Scholar]
  60. Sarchiapone M, Roy A, Camardese G, De Risio S. Further evidence for low serum cholesterol and suicidal behaviour. Journal of Affective Disorders. 2000;61:69–71. doi: 10.1016/s0165-0327(99)00198-6. [DOI] [PubMed] [Google Scholar]
  61. Scanlon SM, Williams DC, Schloss P. Membrane cholesterol modulates serotonin transporter activity. Biochemistry. 2001;40:10507–10513. doi: 10.1021/bi010730z. [DOI] [PubMed] [Google Scholar]
  62. Seefried G, Gumpel K. Low serum cholesterol and triglycerides and risk of death from suicide. Archives of Gerontolology and Geriatrics. 1997;25:111–117. doi: 10.1016/s0167-4943(96)00776-5. [DOI] [PubMed] [Google Scholar]
  63. Sinclair JM, Harriss L, Baldwin DS, King EA. Suicide in depressive disorders: a retrospective case-control study of 127 suicides. Journal of Affective Disorders. 2005;87:107–113. doi: 10.1016/j.jad.2005.03.001. [DOI] [PubMed] [Google Scholar]
  64. Smith GD, Shipley MJ, Marmot MG, Rose G. Plasma cholesterol concentration and mortality. The Whitehall Study. JAMA - Journal of the American Medical Association. 1992;267:70–76. [PubMed] [Google Scholar]
  65. Spitzer RL, Endicott J, Robins E. Research diagnostic criteria: rationale and reliability. Archives of General Psychiatry. 1978;35:773–782. doi: 10.1001/archpsyc.1978.01770300115013. [DOI] [PubMed] [Google Scholar]
  66. Sullivan PF, Joyce PR, Bulik CM, Mulder RT, Oakley-Browne M. Total cholesterol and suicidality in depression. Biological Psychiatry. 1994;36:472–477. doi: 10.1016/0006-3223(94)90643-2. [DOI] [PubMed] [Google Scholar]
  67. Takei N, Kunugi H, Nanko S, Aoki H, Iyo R, Kazamatsuri H. Low serum cholesterol and suicide attempts. British Journal of Psychiatry. 1994;164:702–703. doi: 10.1192/s0007125000034632. [DOI] [PubMed] [Google Scholar]
  68. Tamosiunas A, Reklaitiene R, Radisauskas R, Jureniene K. Prognosis of risk factors and trends in mortality from external causes among middle-aged men in Lithuania. Scandinavian Journal of Public Health. 2005;33:190–196. doi: 10.1080/14034940510005707. [DOI] [PubMed] [Google Scholar]
  69. Tanskanen A, Vartiainen E, Tuomilehto J, Viinamaki H, Lehtonen J, Puska P. High serum cholesterol and risk of suicide. American Journal of Psychiatry. 2000;157:648–650. doi: 10.1176/appi.ajp.157.4.648. [DOI] [PubMed] [Google Scholar]
  70. Tripodianakis J, Markianos M, Sarantidis D, Agouridaki M. Biogenic amine turnover and serum cholesterol in suicide attempt. European Archives of Psychiatry and Clinical Neuroscience. 2002;252:38–43. doi: 10.1007/s004060200007. [DOI] [PubMed] [Google Scholar]
  71. Tsai SY, Kuo CJ, Chen CC, Lee HC. Risk factors for completed suicide in bipolar disorder. Journal of Clinical Psychiatry. 2002;63:469–476. doi: 10.4088/jcp.v63n0602. [DOI] [PubMed] [Google Scholar]
  72. Vevera J, Zukov I, Morcinek T, Papezova H. Cholesterol concentrations in violent and non-violent women suicide attempters. European Psychiatry. 2003;18:23–27. doi: 10.1016/s0924-9338(02)00011-1. [DOI] [PubMed] [Google Scholar]
  73. Young EA, Coryell W. Suicide and the hypothalamic-pituitary-adrenal axis. Lancet. 2005;366:959–961. doi: 10.1016/S0140-6736(05)67348-5. [DOI] [PubMed] [Google Scholar]
  74. Zhang J, McKeown RE, Hussey JR, Thompson SJ, Woods JR, Ainsworth BE. Low HDL cholesterol is associated with suicide attempt among young healthy women: the Third National Health and Nutrition Examination Survey. Journal of Affective Disorders. 2005;89:25–33. doi: 10.1016/j.jad.2005.05.021. [DOI] [PubMed] [Google Scholar]
  75. Zureik M, Courbon D, Ducimetiere P. Serum cholesterol concentration and death from suicide in men: Paris prospective study I. British Medical Journal. 1996;313:649–651. doi: 10.1136/bmj.313.7058.649. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES