Improving Suicide Prevention Through Evidence-Based Strategies: A Systematic Review

SSRIs appeared more effective than noradrenergic drugs for suicidal ideation.

“SSRIs target serotonin, and appeared more effective than noradrenergic drugs for suicidal ideation.(59, 61)”

• cites [59] sourcePartly supported

  The Discussion (page not provided) reports an exploratory finding that patients with more severe baseline suicidal ideation improved more on the SSRI paroxetine than on bupropion. It also says the primary treatment effects were not significant. The claim omits those limits and generalizes a comparison of two drugs to SSRIs and noradrenergic drugs as classes. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [61] sourceDisputed

  The Discussion reports that suicidal ideation decreased with both escitalopram (an SSRI) and nortriptyline, with no overall difference between them. It identifies a narrower exception: men taking nortriptyline had higher rates of treatment-emergent or worsening suicidal ideation. That does not support the unqualified claim that SSRIs appeared more effective. (LLM, conf=0.88)

  Checked by a reasoning model — may be wrong.

  Experimental — accuracy not yet measured

Most subsequent meta-analyses and epidemiological studies in pediatric and adult populations reported a more favorable risk-benefit ratio than FDA analyses, including benefits for suicidal behavior.

“One systematic review of antidepressants in pediatric populations found SSRIs increased odds of a suicide attempt,(53) but most subsequent meta-analyses and epidemiological studies in both pediatric(48),(50) and adult populations(49, 52, 66, 67, 86) reported a more favorable risk-benefit ratio than the FDA analyses,(51, 87) including benefits for suicidal behavior.”

• cites [48] sourceDisputed

  The Results (Suicide-related outcomes; page unavailable) found no significant difference in suicide-related risk between SNRIs and placebo in children and adolescents, not a benefit for suicidal behavior. The Results (Acceptability outcomes; page unavailable) also report significantly more discontinuations for adverse effects with SNRIs. This paper does not support the favorable pediatric risk-benefit finding attributed to it; it does not assess adult populations or compare its findings directly with the FDA analyses. (LLM, conf=0.88)

  Checked by a reasoning model — may be wrong.

  Experimental — accuracy not yet measured
• cites [50] sourceNot checked — no full text of the source
• cites [49] sourcePartly supported

  The Discussion reports antidepressant benefits in youth and adults and says the findings should prompt reconsideration of the risk–benefit equation behind the warning about suicidal thinking in children. It does not report a benefit for suicidal behavior or establish what most subsequent meta-analyses and epidemiological studies found; this paper analyzes trial data for fluoxetine and venlafaxine. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [52] sourcePartly supported

  The COMMENT section reports significantly lower suicide-risk measures with antidepressant treatment in adult and geriatric patients. But it says those adult findings largely agree with the FDA’s, rather than showing a more favorable result, and that the study could not analyze suicidal behavior separately from ideation. It therefore supports an adult suicide-risk benefit, not the full comparison or a distinct benefit for suicidal behavior. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [66] sourceNot checked — no full text of the source
• cites [67] sourceNot checked — no full text of the source
• cites [86] sourceNot checked — no full text of the source

Both pediatric and adult studies of antidepressant effects on suicidal behavior show reductions in suicidal behavior in meta-analyses more readily than in individual RCTs.

“Both pediatric and adult studies of antidepressant effects on suicidal behavior show reductions in suicidal behavior in meta-analyses, more readily than in individual RCTs, probably because the low base rate of suicide attempts requires larger sample sizes.”

• cites [48] sourceDisputed

  The Results (Suicide-related outcomes; page unavailable) report no significant reduction in suicide-related outcomes with SNRIs versus placebo in the pooled pediatric trials (RR=1.09; 95% CI=0.60–1.99). The paper does not examine adult trials, so it cannot support that part of the claim either. (LLM, conf=0.88)

  Checked by a reasoning model — may be wrong.

  Experimental — accuracy not yet measured
• cites [49] sourceCould not verify

  The Discussion (page unavailable) reports improved depression response and remission in youth and adults, but does not report reductions in suicidal behavior. A separate Discussion passage mentions the pediatric warning about suicidal thinking; neither it nor the Methods passages establish the claimed comparison between meta-analyses and individual RCTs. Searches for suicidal behavior, suicide attempts, and the proposed low-base-rate explanation found no supporting passage. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [50] sourceNot checked — no full text of the source
• cites [52] sourcePartly supported

  The COMMENT reports significant reductions in suicide-risk measures for treated adults in an analysis combining trial data, and the METHODS notes that suicide attempts and deaths were too few for well-powered comparisons. But the RESULTS > YOUTH FLUOXETINE STUDIES found no significant treatment effect in youths. The paper also says it lacked power to analyze suicidal behavior separately from ideation, so it does not establish the claimed reduction in suicidal behavior for both age groups or directly compare meta-analyses with individual RCTs. Page numbers are unavailable in our copy. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [53] sourceDisputed

  The Results report the opposite direction for the paper’s pooled analysis: SSRIs significantly increased suicide attempts compared with placebo, rather than reducing them. The Introduction notes that rare events can make differences hard to detect in individual studies, but its mention of an earlier reduction concerns suicidal ideation, not suicidal behavior. (LLM, conf=0.91)

  Checked by a reasoning model — may be wrong.

  Experimental — accuracy not yet measured
• cites [54] sourceNot checked — no full text of the source
• cites [66] sourceNot checked — no full text of the source
• cites [67] sourceNot checked — no full text of the source
• cites [86] sourceNot checked — no full text of the source

The U.K. study observed the decline in a sample of 19,248 suicides who were in treatment within 12 months of death studied 1997-2012.

“The decline in suicide rates was observed in the study sample of 19,248 suicides who were in treatment within 12 months of death studied 1997–2012.”

• cites [34] sourceSupported

Targeting high school students prevented student suicide attempts.

“Targeting high school students prevented student suicide attempts (29, 30) whereas targeting teachers (29) and all but one parent study (Table S1), did not find benefit.”

• cites [29] sourceNot checked — no full text of the source
• cites [30] sourceSupported

A U.K. study found reduced suicide rates with incident rate ratios of .71-.79 for an array of clinical service changes including improved depression management, continuity of care, improved community services, and lower staff turnover.

“One study in the United Kingdom (34) found benefit for an array of clinical service changes including improved depression management, continuity of care from adolescent patient services to adult patient services, improved community services and lower staff turnover and found reduced suicide rates with incident rate ratios of .71–.79.”

• cites [34] sourcePartly supported

  The Summary (p. 1) reports incidence rate ratios of 0.71–0.79 for service changes, and the Results (p. 5) identify depression guidance, community treatment teams, and transitions to adult services among changes associated with lower suicide rates. The paper treats low staff turnover separately as an organisational factor that strengthened the association between service changes and reduced suicide rates (Results, p. 6), not as a clinical service change with an incidence rate ratio in that range. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Training primary care doctors and nurses to better screen and treat depression, with supplemental psychiatric help, lowered nonfatal suicide attempts.

“Training primary care doctors and nurses at the local and state level to better screen and treat depression with supplemental help available from psychiatrists, lowered suicide rates,(12–19) nonfatal suicide attempts(20, 21) and ideation(22) (Table 2).”

• cites [20] sourcePartly supported

  The Results (General effect on suicidal acts; page unavailable) report fewer attempted suicides relative to controls in Portugal, but no overall reduction across the four countries. The implementation findings describe GP training, not a demonstrated effect of training doctors and nurses with supplemental psychiatric help; the Discussion says the contribution of any single intervention measure cannot be isolated. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [21] sourceNot checked — no full text of the source

In a trial of 4,133 high school students, a mental health and suicide education program produced less suicidal behavior than control over the next year.

“In the second study, 4,133 high students were randomized to a mental health and suicide education program or control group. Less suicidal behavior was observed in the active intervention group relative to the control group over the next year.”

• cites [30] sourcePartly supported

  The Abstract (Methods) reports 4,133 high school students assigned to intervention and control groups, and the Discussion reports significantly fewer suicide attempts in the intervention group. But the outcome was measured approximately three months after the program, not over the next year. (LLM, conf=0.92)

  Experimental — accuracy not yet measured

Primary care and non-psychiatric settings see about double the rate of mental health professionals among future suicide decedents.

“Doctors in primary care and other non-psychiatric care settings see 45% of future suicide decedents in the 30 days prior to suicide, and 77% within 12 months of suicide,(11) about double the rate consulting mental health professionals.”

• cites [11] sourcePartly supported

  The Results report approximately 45% contact with primary care providers within one month and 77% within one year, versus 19% and 32% contact with mental health services—roughly twice as high. These figures concern primary care providers; the paper does not establish the same rates for other non-psychiatric care settings. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

Repeated education sessions reduced suicide rates progressively for years.

“Repeated education sessions reduced suicide rates progressively for years.(12, 21)”

• cites [12] sourcePartly supported

  The Intervention Training section documents repeated education sessions, and the Primary Outcome section reports a lower mean annual suicide rate during the five intervention years than during the five preceding years. The paper does not establish that rates fell progressively year by year; the Comment section also says it cannot determine whether education, additional resources, or both accounted for the change. (LLM, conf=0.89)

  Experimental — accuracy not yet measured
• cites [21] sourceNot checked — no full text of the source

In a study of 40 primary care practices in the Netherlands, a drop in depression recognition from 65% to 44% was associated with higher suicide attempt rates in men.

“Conversely, a study of 40 primary care medical practices in the Netherlands found a drop in depression recognition from 65% to 44% was associated with higher suicide attempt rates in men.(28)”

• cites [28] sourcePartly supported

  The Abstract’s Setting identifies 40 Dutch general practices, and its Results report both a rise in male suicide attempts and a decline from 65% to 44% in reported depression history. But those percentages concern all patients who died by suicide, not men who attempted suicide. The Results section on suicidal behaviour after 2007 gives different figures for male suicide decedents (60% to 46%); the paper does not establish an association between declining depression recognition and male attempt rates. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

A single-day training session produced no benefit for suicide deaths over 3 years.

“In contrast, a single day training session produced no benefit for suicide deaths over 3-years.(27)”

• cites [27] sourcePartly supported

  The Education program section describes a four-hour training program, and Table 4 compares suicide deaths over the three years before and after it. The Results report a 12% decrease in intervention regions versus 4% in the control region, but the difference was not statistically significant. Thus the paper supports no demonstrated benefit for suicide deaths, not the categorical claim that the training produced no benefit. (LLM, conf=0.87)

  Experimental — accuracy not yet measured

Training primary care doctors and nurses to better screen and treat depression, with supplemental psychiatric help, lowered suicide rates.

“Training primary care doctors and nurses at the local and state level to better screen and treat depression with supplemental help available from psychiatrists, lowered suicide rates,(12–19) nonfatal suicide attempts(20, 21) and ideation(22) (Table 2).”

• cites [12] sourceSupported
• cites [13] sourceNot checked — no full text of the source
• cites [14] sourcePartly supported

  The Introduction (page not specified) describes depression screening by trained public health nurses, and the Discussion reports a 74% reduction in suicide risk among elderly women in the intervention area, but not among men. The intervention also included psychoeducation and group activities; the Discussion says it cannot determine which component accounted for the reduction. The paper does not describe training primary care doctors to screen and treat depression, although it records occasional referrals to psychiatrists. (LLM, conf=0.89)

  Experimental — accuracy not yet measured
• cites [15] sourcePartly supported

  The Discussion reports that the suicide rate fell among elderly females following the intervention. But it describes the intervention as group activities, psychoeducation, and self-assessment of depression—not training primary care doctors and nurses to screen and treat depression. The Self-assessment section explicitly says the questionnaire was used to recommend assessment, “but not to screen.” (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [16] sourceNot checked — no full text of the source
• cites [17] sourceNot checked — no full text of the source
• cites [18] sourceNot checked — no full text of the source
• cites [19] sourceNot checked — no full text of the source

Screening for suicide risk coupled with effective referral for evaluation and treatment was found by some studies to prevent suicidal behavior.

“Screening for suicide risk to identify otherwise undetected at-risk individuals,(30, 37) if coupled with effective referral for evaluation and treatment, was found by some studies to prevent suicidal behavior.(15–17, 37, 38)”

• cites [15] sourcePartly supported

  The Introduction (page unavailable) reports that earlier programs combining depression screening with psychiatric follow-up reduced suicide rates among older adults. It does not establish that screening specifically for suicide risk, coupled with effective referral, produced that result; the paper says its own group intervention was not a screening program. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [16] sourceNot checked — no full text of the source
• cites [17] sourceNot checked — no full text of the source
• cites [37] Dube P, 2010, Prim Care Companion J Clin Psychiatry sourcePartly supported

  The Discussion (page unavailable) recommends asking about suicidal ideation and describes assessment that can guide treatment, follow-up, or referral. But it says the degree to which suicide is preventable remains uncertain; it does not report that studies found screening coupled with referral prevented suicidal behavior. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [38] sourceCould not verify

  The Methods and Clinical implications sections describe screening and limited follow-up assistance, but do not evaluate screening coupled with effective referral and treatment as a way to prevent suicidal behavior. The Strengths and limitations section says the trial lacked sufficient power to detect differences in self-harm; its conclusion concerns whether screening induced suicidal thoughts, not whether it prevented suicidal behavior. (LLM, conf=0.76)

  Experimental — accuracy not yet measured

In areas with suicide rates 3-10 times higher than the U.S., screening for depression combined with referral for depression treatment by a primary care doctor or psychiatrist lowered suicide rates relative to a contemporaneous geographic control.

“Areas with suicide rates 3–10 times higher than the U.S., have shown that screening for depression combined with referral for depression treatment by a primary care doctor or psychiatrist lowered suicide rates relative to a contemporaneous geographic control.(12, 15–17)”

• cites [12] sourcePartly supported

  The Improving Access to Depression Treatment section describes GP depression screening and referral to a depression clinic. The Primary Outcome section reports a greater decline in suicide rates than in the surrounding county, but no overall difference in the change relative to the nearby local control region; the rural-women comparison was a trend (P=.07). The claim overstates the result if “contemporaneous geographic control” means that local control region. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [15] sourcePartly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured
• cites [16] sourceNot checked — no full text of the source
• cites [17] sourceNot checked — no full text of the source

The relationship between suicide deaths and nonfatal suicide attempts is closer than the relationship between suicidal ideation and suicide deaths.

“We focused on suicidal behavior as an outcome and not suicidal ideation, because there is closer a relationship between suicide deaths and nonfatal suicide attempts than there is between suicidal ideation and suicide deaths.(7)”

• cites [7] sourceCould not verify

  The Prevalence and Characteristics of suicidal behaviours as risk factors sections report attempts among people with suicidal ideation and transitions from ideation to attempts. The Background calls ideation, plans and attempts precursors to suicide, but no passage compares how closely attempts and ideation are related to suicide deaths (searched: suicide deaths, nonfatal attempts, completed suicide). (LLM, conf=0.76)

  Experimental — accuracy not yet measured

Antidepressants reduced suicidal ideation in 9 of 12 studies that reported effects on suicidal ideation.

“More promisingly, antidepressants reduced suicidal ideation in 9/12 studies that reported effects on suicidal ideation.(55, 57, 59–65)”

• cites [55] sourceNot checked — no full text of the source
• cites [57] sourceNot checked — no full text of the source
• cites [59] sourcePartly supported

  The Discussion reports exploratory evidence of greater improvement in suicidal ideation with paroxetine than bupropion among patients with more severe baseline ideation. The Introduction summarizes 10 studies comparing antidepressant classes, but neither section reports that antidepressants reduced ideation in 9 of 12 studies. No page numbers are available in our copy. (LLM, conf=0.82)

  Experimental — accuracy not yet measured
• cites [60] sourceNot checked — no full text of the source
• cites [61] sourcePartly supported

  The Discussion and Conclusion (pages unavailable in our copy) report that suicidal ideation decreased during treatment with escitalopram or nortriptyline in the GENDEP trial. This supports the paper as one example of a study finding a reduction, but it does not establish the citing sentence’s count of 9 out of 12 studies. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [62] sourceNot checked — no full text of the source
• cites [63] sourcePartly supported

  The Efficacy Results (p. 625) report reduced suicidal thoughts with esketamine versus placebo at 4 hours, so this paper supports one instance of the claimed effect. It is a single trial, however, and does not report that antidepressants reduced suicidal ideation in 9 of 12 studies. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [64] sourceNot checked — no full text of the source
• cites [65] sourceNot checked — no full text of the source

Ketamine reduced suicidal ideation within 1-4 hours in major depression.

“Ketamine, an NMDA glutamate receptor antagonist, reduced suicidal ideation within 1–4 hours in major depression(71–75) and bipolar disorder.(76, 77)”

• cites [71] sourcePartly supported

  The Introduction reports earlier findings suggesting that ketamine can reduce suicidal ideation within 40 minutes, including in treatment-resistant depression. This trial studied patients with treatment-resistant unipolar major depression but measured its suicidal-ideation outcome at 24 hours, not at 1–4 hours; the claimed interval is therefore not established by this trial. (LLM, conf=0.80)

  Experimental — accuracy not yet measured
• cites [72] sourceSupported
• cites [73] sourcePartly supported

  The Results section reports significantly lower suicidal-ideation scores after ketamine, but the measurements were on days 1 and 3 in patients with newly diagnosed cancer. The Discussion mentions antidepressant effects as soon as 40 minutes in depressed patients; it does not establish reduced suicidal ideation within 1–4 hours in major depression. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [74] sourcePartly supported

  The Suicidal ideation results (section 3.2; page unavailable) report that ketamine alleviated suicidal ideation 90–180 minutes after infusion, within the claimed 1–4-hour window. But the Participants section (2.1; page unavailable) includes major depression alongside bipolar depression, depression NOS, and dysthymia; the reported effect is for the combined group, not specifically for patients with major depression. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites [75] sourceNot checked — no full text of the source

Screening in the U.S. military indicated current ideation added predictive power to a history of a previous suicide attempt.

“Screening in the U.S. military(42) indicated current ideation added predictive power to a history of a previous suicide attempt.”

• cites [42] sourcePartly supported

  The Results table lists both prior suicide attempts and the frequency of outpatient visits recording suicidal ideation as predictors in a model for soldiers with prior psychiatric hospitalization. It does not report that screening for current ideation added predictive power beyond a history of attempts; its ideation measures concern visits in the preceding three or twelve months. The Abstract says clinician-evaluated suicide-risk information might improve a future model. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

A meta-analysis of non-randomized studies suggested risk of attempts and suicide was 5 times less with lithium treatment.

“A meta-analysis of RCTs in bipolar disorder found no data on lithium’s effect on suicide prevention(81), however a meta-analysis of non-randomized studies suggested that risk of attempts and suicide was 5 times less with lithium treatment.(82)”

• cites [82] sourcePartly supported

  The Meta-analysis: all suicide acts section reports a 4.91-fold lower risk of suicide or attempts during lithium treatment, supporting the approximate five-fold figure. But that estimate pools 31 studies, including five RCTs, so describing it as a meta-analysis of non-randomized studies alone is inaccurate. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

Ketamine reduced suicidal ideation within 1-4 hours in bipolar disorder.

“Ketamine, an NMDA glutamate receptor antagonist, reduced suicidal ideation within 1–4 hours in major depression(71–75) and bipolar disorder.(76, 77)”

• cites [76] sourceSupported
• cites [77] sourceNot checked — no full text of the source

More complex EMR-based screening, bypassing gatekeepers, may improve identification of higher risk patients.

“More complex EMR-based screening, bypassing gatekeepers, may improve identification of higher risk patients.(43–47)”

• cites [43] sourcePartly supported

  The Historical Administrative Data Study section (page unavailable) describes using soldiers’ administrative records to study predictors of suicide risk, and the Discussion (page unavailable) proposes adding more detailed predictors. The paper does not evaluate EMR-based screening, show that it improves identification of high-risk patients, or discuss bypassing gatekeepers. (LLM, conf=0.79)

  Experimental — accuracy not yet measured
• cites [44] sourceNot checked — no full text of the source
• cites [45] sourcePartly supported

  The Discussion (page not provided) reports that an algorithm using Army/DoD administrative data identified a highest-risk group containing 52.9% of post-hospital suicides. It does not say the approach bypasses gatekeepers; instead, it describes the algorithm as a component of, not a substitute for, clinical assessment. (LLM, conf=0.85)

  Experimental — accuracy not yet measured
• cites [46] sourcePartly supported

  The Discussion (page unavailable) says that considering additional patient characteristics would improve risk prediction and help identify patients at highest risk. The Methods (page unavailable) describe using electronic health records, but the study does not say that screening bypasses gatekeepers or evaluate that approach. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites [47] sourcePartly supported

  The Introduction (Background and Significance; page not provided) describes EHR-based machine learning as a way to address limitations of clinician-administered risk assessments. The Discussion (Conclusions; page not provided) says better algorithms could improve follow-up for patients at risk, but it does not establish that screening bypasses clinical gatekeepers or improves identification in practice. (LLM, conf=0.83)

  Experimental — accuracy not yet measured

Brief screening tools like the P4 or C-SSRS may improve triage.

“Brief screening tools like the P4, that assesses the 4P’s (past suicide attempt, suicide plan, probability of completing suicide and preventive factors), or the Columbia Suicide Severity Rating Scale (C-SSRS), that assesses severity of previous suicidal behavior and current suicidal ideation, may improve triage.(37, 39)”

• cites [37] Dube P, 2010, Prim Care Companion J Clin Psychiatry sourceSupported
• cites [39] sourceSupported

A meta-analysis of all FDA-registered randomized placebo-controlled studies of fluoxetine and venlafaxine found these medications decreased suicidal ideation and behavior.

“A meta-analysis of all FDA-registered randomized placebo-controlled studies of fluoxetine and venlafaxine found these medications decreased suicidal ideation and behavior.(49, 52)”

• cites [49] sourcePartly supported

  The Discussion (page unavailable) says the paper analyzed sponsor-conducted, placebo-controlled trials of fluoxetine and venlafaxine. Its reported benefits concern depressive symptoms, response, and remission—not a decrease in suicidal ideation or behavior. The passage also does not establish that the trials were all FDA-registered. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [52] sourcePartly supported

  The Study Selection section describes sponsor-conducted randomized trials of fluoxetine and venlafaxine, and the Conclusions report decreased suicidal thoughts and behavior in adult and geriatric patients. The claim omits the important qualifier that no significant treatment effect was found in youths. Our copy also describes the trials as sponsor-conducted, rather than establishing that they were all FDA-registered. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

The C-SSRS predicted suicide attempts with an OR of 4.8 (95% CI 2.23-10.32, p<.001) in adolescents and young adults following an emergency psychiatric evaluation.

“The C-SSRS predicted suicide attempts with an OR of 4.8 (95% CI 2.23–10.32, p<.001) in adolescents and young adults following an emergency psychiatric evaluation.(41)”

• cites [41] sourcePartly supported

  The Discussion says several C-SSRS items predicted future attempts in adolescents and emerging adults seeking psychiatric emergency services. But Table 3, Model 1 (Future Attempter Index Visit Clinical Characteristics; page unavailable) assigns OR 4.80 (95% CI 2.23–10.32, p<.001) specifically to a lifetime history of suicide attempt, not to the C-SSRS overall. (LLM, conf=0.91)

  Experimental — accuracy not yet measured

Two of four RCTs showed lithium benefit for suicidal behavior compared with various other medications.

“Two of four RCTs showed benefit for suicidal behavior with lithium compared with various other medications.(55, 58, 79, 80)”

• cites [55] sourceNot checked — no full text of the source
• cites [58] sourceNot checked — no full text of the source
• cites [79] sourcePartly supported

  The Methods and Lithium efficacy sections identify this as a randomized trial of lithium plus usual care versus usual care alone, with self-harm and suicide outcomes. The Lithium efficacy section reports no significant difference between arms (p = 0.676). This paper can serve as an example of a relevant RCT, but it does not establish that two of four RCTs showed benefit or that its comparator was another medication. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [80] sourcePartly supported

  The Conclusions report that this paper is an RCT comparing lithium with valproate and found no difference in suicidal outcomes, so it can serve as one of the trials counted. Our copy does not establish that two of four RCTs found a benefit for lithium over other medications; the Discussion’s reference to three of four studies concerns comparisons using other methodologies. (LLM, conf=0.83)

  Experimental — accuracy not yet measured

SSRIs may not increase treatment-emergent suicidal ideation or behavior, even in youth, contrary to earlier FDA findings.

“Contrary to earlier FDA findings, SSRIs may work without increasing risk of treatment-emergent suicidal ideation or behavior, even in youth.(67–69)”

• cites [67] sourceNot checked — no full text of the source
• cites [68] sourcePartly supported

  The Results and Discussion report no significant increase in suicidal ideation or behavior with vilazodone versus placebo among younger adults aged 18–24. That supports a narrower version of the claim, but the paper does not establish the same result for SSRIs generally or for children and adolescents; its Introduction recounts the FDA finding of increased risk in pediatric patients. Page numbers are not available in our copy. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [69] sourcePartly supported

  In “Suicidality in levomilnacipran ER MDD trials” (p. 479), the authors report no greater risk of suicidal ideation or behavior with levomilnacipran ER than placebo among younger adults, while cautioning that the subgroup was small. This supports the possibility that an antidepressant need not increase that risk, but the trials studied levomilnacipran ER, not SSRIs, and enrolled adults aged 18–80, not children or adolescents. The Introduction also describes the FDA warning based on pediatric and young-adult trials rather than establishing that the warning was wrong. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Internet CBT, with or without phone follow-up, was not more effective than waitlist controls for reducing suicidal ideation.

“Internet CBT, with or without phone follow-up, was not more effective compared with waitlist controls for reducing suicidal ideation.(121, 125)”

• cites [121] sourcePartly supported

  The Discussion reports that suicidal ideation declined across intervention conditions, with no consistent overall effect by condition. However, the Method identifies the comparator as treatment as usual—participants could contact the helpline if needed—not a waitlist control. (LLM, conf=0.90)

  Experimental — accuracy not yet measured
• cites [125] sourcePartly supported

  The Results (Original article; page unavailable) report no statistically significant difference in suicidal ideation between Reframe-IT and the control group (p=0.593). But the Methods (Abstract and interventions; page unavailable) identify the comparison as internet CBT plus treatment as usual versus treatment as usual alone, not a waitlist; the paper does not test a separate phone-follow-up condition. (LLM, conf=0.90)

  Experimental — accuracy not yet measured

Safety planning interventions administered in the emergency department with follow-up telephone contact produced a 45% reduction in suicidal behaviors compared with treatment as usual.

“A cohort comparison study of safety planning interventions, administered in ED with follow-up telephone contact produced a 45% reduction in suicidal behaviors vs. TAU.(115)”

• cites [115] sourcePartly supported

  The Suicidal Behavior Report Outcomes section reports 45% fewer suicidal behaviors over six months among patients receiving ED safety planning with telephone follow-up than among those receiving usual care. The paper describes this as an association in a cohort comparison study; saying the intervention “produced” the reduction implies a causal conclusion the study does not establish. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

The greatest risk of suicidal behavior is after discharge from the emergency room or hospital inpatient units.

“The greatest risk of suicidal behavior is after discharge from the emergency room (ER) or hospital inpatient units.(110–112)”

• cites [110] sourceNot checked — no full text of the source
• cites [111] sourceCould not verify

  The Treatment Modalities and Discussion passages address depression treatment and psychological distress, not suicidal behavior after discharge. Searches for suicide, emergency-room discharge, and inpatient discharge found no passage in our copy addressing when suicidal-behavior risk is greatest; page numbers were unavailable. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [112] sourcePartly supported

  The Results (page not supplied) report that 27 of 33 survivors who later died by suicide did so within a year of their initial attempt. The Conclusions also discuss protective follow-up appointments scheduled at discharge from an emergency department or inpatient service. The paper does not establish that risk of suicidal behavior is greatest immediately after discharge; its timing finding concerns deaths after an initial attempt, not time since discharge or all suicidal behavior. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

Sending caring text messages over one year to active military personnel who had reported a suicide attempt or suicidal ideation lowered subsequent suicide attempts by almost half.

“Another study used a similar approach by sending caring text messages over one year to active military personnel who had reported a suicide attempt (50%) or suicidal ideation, but not in the context of discharge from hospital or ED. The intervention lowered subsequent suicide attempts by almost half.(114)”

• cites [114] sourcePartly supported

  The Primary Outcomes section (page unavailable) reports 48% lower odds of at least one suicide attempt during follow-up with caring texts (9.0% vs 14.9%), supporting the main finding. The Participant Characteristics section reports that 44.3%, not 50%, had a lifetime suicide attempt and that 53.0% were referred from an inpatient unit. The paper does not establish the citing sentence’s categorical description that the intervention was outside the context of hospital discharge. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

Faced with no access to their chosen method, most individuals do not turn to alternative means for suicide, as shown for coal and gas or firearm restriction.

“Surprisingly, faced with no access to their chosen method, most individuals do not turn to alternative means for suicide, as shown for coal and gas(156) or firearm restriction.(127)”

• cites [156] Kreitman N, 1976, Br J Prev Soc Med sourcePartly supported

  The ‘Domestic Gas and Other Modes of Death in Suicide’ section reports that the fall in coal-gas suicides was large enough to lower total suicide rates despite increases in deaths by other means in some groups. But ‘Some Implications’ treats whether individuals switched methods as an open question and says virtually nothing is known about it. The paper therefore supports an aggregate decline, not the claim that most individuals did not turn to alternative means; the firearm citation is separate. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites [127] sourceNot checked — no full text of the source

Screening for suicidal ideation does not induce subsequent suicidal ideation.

“Although it does not induce subsequent suicidal ideation,(38, 147) screening a non-help-seeking population for suicidal ideation and nonfatal suicide attempts, beyond screening for major depression, remains a debated approach.(40)”

• cites [38] sourcePartly supported

  The Conclusions report that screening among primary-care patients with signs of depression “does not appear to induce” feelings that life is not worth living. The Clinical implications also report no apparent increase in subsequent suicidal ideation over the short follow-up. The citing clause states this more definitively and without the study’s population and time limits; the nearby discussion of a non-help-seeking population is not established by this trial. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites [147] sourceNot checked — no full text of the source

Ketamine reduces suicidal ideation within hours.

“Ketamine reduces suicidal ideation in hours but is untested for suicidal behavior prevention.”

Education of gatekeepers about youth suicidal behavior lacks effectiveness.

“Education of gatekeepers about youth suicidal behavior lacks effectiveness.”

Educating youth about depression and suicidal behavior prevents suicidal behavior.

“Educating youth regarding depression and suicidal behavior, and active outreach of psychiatric patients after discharge or a suicidal crisis, prevents suicidal behavior.”

Active outreach to psychiatric patients after discharge or a suicidal crisis prevents suicidal behavior.

“Educating youth regarding depression and suicidal behavior, and active outreach of psychiatric patients after discharge or a suicidal crisis, prevents suicidal behavior.”

Training primary care physicians in depression recognition and treatment prevents suicide.

“Training primary care physicians in depression recognition and treatment prevents suicide.”

Meta-analyses find that antidepressants prevent suicide attempts.

“Meta-analyses find antidepressants prevent suicide attempts, but individual RCTs appear under-powered.”

Active screening for suicidal ideation or behavior is not proven to be better than screening for depression alone.

“Active screening for suicidal ideation or behavior is not proven to be better than just screening for depression.”

No randomized controlled trials are reported for gatekeeper training for adult suicidal behavior prevention.

“No RCTs are reported for gatekeeper training for adult suicidal behavior prevention.”

Ketamine is untested for preventing suicidal behavior.

“Ketamine reduces suicidal ideation in hours but is untested for suicidal behavior prevention.”

In a trial of 168 high schools and 11,110 students, the YAM student education program prevented suicidal behavior relative to control.

“YAM prevented suicidal behavior relative to the control condition, but the screening intervention and educating teachers/staff gatekeepers did not.”

• cites [29] sourceNot checked — no full text of the source

The Garrett Lee Smith benefit faded once programming stopped regardless of years implemented.

“Follow-up found the benefit faded once programming stopped regardless of years it was implemented.”

• cites [32] sourceNot checked — no full text of the source

In 1,126 U.S. counties, Garrett Lee Smith Act interventions directed at young persons were associated with a decline in youth suicide rates compared with 969 comparable counties.

“The intervention counties exhibited a decline in youth suicide rates that did not extend to other causes of death in youth or to adult suicide rates, compared with 969 demographical and sociologically comparable counties where this program was not implemented.”

• cites [32] sourceNot checked — no full text of the source

All but one parent study did not find benefit for youth suicide prevention.

“Targeting high school students prevented student suicide attempts (29, 30) whereas targeting teachers (29) and all but one parent study (Table S1), did not find benefit.”

Training primary care doctors and nurses to better screen and treat depression, with supplemental psychiatric help, lowered suicidal ideation.

“Training primary care doctors and nurses at the local and state level to better screen and treat depression with supplemental help available from psychiatrists, lowered suicide rates,(12–19) nonfatal suicide attempts(20, 21) and ideation(22) (Table 2).”

• cites [22] sourceNot checked — no full text of the source

Algorithm-driven electronic health record screening, internet-based screening, and smartphone passive monitoring to identify high-risk patients are under-studied.

“Algorithm-driven electronic health records (EHR) screening, internet-based screening and smartphone passive monitoring to identify high-risk patients are under-studied.”

Means restriction, including restriction of firearms, prevents suicide.

“Means restriction, including firearms, prevents suicide, but is rarely employed in the U.S., even though firearms are used in half of all U.S. suicides.”

• AssumptionRests on an assumption: Firearms account for half of all suicides in the United States.

Depression-management education of doctors lowers adult suicide rates.

“Depression-management education of doctors lowers adult suicide rates (Table 2), but extending the education to other gatekeepers and the general public is not proven to further lower nonfatal and fatal suicide attempt rates.(31)”

Targeting teachers did not prevent student suicide attempts.

“Targeting high school students prevented student suicide attempts (29, 30) whereas targeting teachers (29) and all but one parent study (Table S1), did not find benefit.”

• cites [29] sourceNot checked — no full text of the source

The poorest districts where the U.K. interventions were deployed had the greatest declines in suicide rates.

“This study did not control for changes in the secular suicide rate, and there was no randomization of the intervention, despite the fact that the poorest districts where the interventions were deployed had the greatest declines in suicide rates.(35)”

• cites [35] sourceNot checked — no full text of the source

In the same trial, the professional screening program with referral did not prevent suicidal behavior.

“YAM prevented suicidal behavior relative to the control condition, but the screening intervention and educating teachers/staff gatekeepers did not.”

• cites [29] sourceNot checked — no full text of the source

A U.S. study of staff education and frequent screening of suicide risk in all psychiatric patients in the Henry Ford Health System resulted in 9 consecutive quarters without a suicide compared with 80/100,000 suicides in 2000.

“A U.S. study examined the effect of staff education and frequent screening of suicide risk in all psychiatric patients in the Henry Ford Health System that resulted in 9 consecutive quarters without a suicide compared with 80/100,000 suicides in 2000.”

Extending education to other gatekeepers and the general public is not proven to further lower nonfatal and fatal suicide attempt rates.

“Depression-management education of doctors lowers adult suicide rates (Table 2), but extending the education to other gatekeepers and the general public is not proven to further lower nonfatal and fatal suicide attempt rates.(31)”

• cites [31] sourceNot checked — no full text of the source

In youth prevention studies, 7 of 9 found less suicidal ideation.

“Although 4/9 prevention studies in youth reported less suicidal behavior (Table S1), 7/9 found less ideation (Table S1).”

In the same trial, educating teachers/staff gatekeepers did not prevent suicidal behavior.

“YAM prevented suicidal behavior relative to the control condition, but the screening intervention and educating teachers/staff gatekeepers did not.”

• cites [29] sourceNot checked — no full text of the source

System-wide health care changes have not been tested in a randomized study.

“System-wide health care changes have not been tested in a randomized study but promising results have emerged from opportunistic before and after studies.(33)”

In the Henry Ford program, the suicide rate declined to zero in 2008-2010.

“The program ran from 2001–2007, and the rate declined to zero in the period 2008–2010, inspiring the notion of zero suicide as a goal for health care systems.”

In youth prevention studies, 4 of 9 reported less suicidal behavior.

“Although 4/9 prevention studies in youth reported less suicidal behavior (Table S1), 7/9 found less ideation (Table S1).”

There are no controlled trials of targeted gatekeeper education in adults.

“There are no controlled trials of targeted gatekeeper education in adults.”

Promising results for system-wide health care changes have emerged from opportunistic before-and-after studies.

“System-wide health care changes have not been tested in a randomized study but promising results have emerged from opportunistic before and after studies.(33)”

• cites [33] sourceNot checked — no full text of the source

Eleven of sixteen ecological studies reported lithium levels in drinking water were linked to lower suicide rates.

“Although one study found no correlation between bipolar disorder and many other psychiatric disorders with lithium levels in drinking water (83), 11/16 ecological studies report these levels were linked to lower suicide rates.(84)”

• cites [84] sourceNot checked — no full text of the source

Ketamine's effect on suicidal behavior has never been evaluated.

“Used adjunctively with other psychotropic medications, one RCT found ketamine’s benefit for suicidal ideation persisted for weeks,(72) but its effect on suicidal behavior has never been evaluated.”

Subsequent pharmaco-epidemiological studies found SSRIs reduced suicidal behavior more broadly, including in young adults.

“An FDA meta-analysis found antidepressants lowered risk of suicidal behavior in older age groups, but subsequent pharmaco-epidemiological studies found SSRIs reduced suicidal behavior more broadly, including in young adults.(66, 67)”

• cites [66] sourceNot checked — no full text of the source
• cites [67] sourceNot checked — no full text of the source

The effective medications in those adult studies belonged to pharmacologically diverse classes.

“Medications reduced suicidal behavior in 4/12 adult studies,(55–58) but those effective medications belonged to pharmacologically diverse classes.”

• cites [55] sourceNot checked — no full text of the source
• cites [56] sourceNot checked — no full text of the source
• cites [57] sourceNot checked — no full text of the source
• cites [58] sourceNot checked — no full text of the source

Meta-analyses of RCT data show more impressive benefit for suicidal behavior than individual RCTs.

“Meta-analyses of RCT data show more impressive benefit for suicidal behavior than individual RCTs.”

Medications reduced suicidal behavior in 4 of 12 adult studies.

“Medications reduced suicidal behavior in 4/12 adult studies,(55–58) but those effective medications belonged to pharmacologically diverse classes.”

• cites [55] sourceNot checked — no full text of the source
• cites [56] sourceNot checked — no full text of the source
• cites [57] sourceNot checked — no full text of the source
• cites [58] sourceNot checked — no full text of the source

In adolescents, young adults, and older adults, suicide attempt risk was greatest in the month before antidepressant medication began.

“Longitudinal pharmaco-epidemiologic studies in adolescents, young adults and older adults indicated that the greatest risk for a suicide attempt was in the month before antidepressant medication began, and after the medication was initiated, the risk declined progressively over months.(66, 67, 70)”

• cites [66] sourceNot checked — no full text of the source
• cites [67] sourceNot checked — no full text of the source
• cites [70] sourceNot checked — no full text of the source

An FDA meta-analysis found antidepressants lowered risk of suicidal behavior in older age groups.

“An FDA meta-analysis found antidepressants lowered risk of suicidal behavior in older age groups, but subsequent pharmaco-epidemiological studies found SSRIs reduced suicidal behavior more broadly, including in young adults.(66, 67)”

• cites [51] sourceNot checked — no full text of the source

Other studies did not find screening and referral of at-risk high school students to be effective.

“However, other studies did not finding screening and referral of at-risk high school students to be effective.(29)”

• cites [29] sourceNot checked — no full text of the source

After antidepressant medication initiation, suicide attempt risk declined progressively over months.

“Longitudinal pharmaco-epidemiologic studies in adolescents, young adults and older adults indicated that the greatest risk for a suicide attempt was in the month before antidepressant medication began, and after the medication was initiated, the risk declined progressively over months.(66, 67, 70)”

• cites [66] sourceNot checked — no full text of the source
• cites [67] sourceNot checked — no full text of the source
• cites [70] sourceNot checked — no full text of the source

Psychodynamic psychotherapies for borderline personality disorder prevent suicidal or self-harm behavior in most controlled studies.

“Psychodynamic psychotherapies for BPD prevent suicidal or self-harm behavior in most controlled studies (Table S3a).”

• cites Table S3aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Educating non-psychiatrist primary care physicians to better diagnose and treat major depression prevents suicide and non-fatal suicide attempts.

“Educating non-psychiatrist primary care physicians to better diagnose and treat major depression prevents suicide and non-fatal suicide attempts (Table 2).”

• cites Table 2Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Combinations of pharmacotherapy and psychotherapy showed no advantage for suicidal behavior.

“Given efficacy evidence for pharmacotherapy and psychotherapy separately, combinations of both showed no advantage for suicidal behavior (Table S3b).”

• cites Table S3bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

No brain stimulation therapies have demonstrated suicide behavior prevention.

“No brain stimulation therapies have demonstrated suicide behavior prevention.”

Determination of imminent risk has evolved towards continuous monitoring via mobile devices.

“Determination of imminent risk has evolved towards continuous monitoring via mobile devices.(148, 149)”

• cites [148] sourceNot checked — no full text of the source
• cites [149] sourceNot checked — no full text of the source

Individual randomized controlled trials of antidepressants appear under-powered.

“Meta-analyses find antidepressants prevent suicide attempts, but individual RCTs appear under-powered.”

The population targeted for education appears to be a key factor in youth suicide prevention.

“A key factor appears to be the population targeted for education.”

Suicide risk can be understood in terms of a stress diathesis model.

“Suicide risk can be understood in terms of a stress diathesis model (8) (Figure 1) where stress results from an internal stressor, usually a psychiatric illness, present in about 90% of all suicides and most commonly major depression, and/or an external stressor involving life events.”

• cites [8] sourceNot checked — no full text of the source

The diathesis moderates suicide risk.

“The diathesis moderates suicide risk, and that risk can be influenced by acute alcohol or drug abuse adversely via mood or disinhibition (Figure 1).”

• cites [8] sourceNot checked — no full text of the source

Educating non-psychiatrist physicians may prevent far more suicides than further training for psychiatrists.

“Therefore, educating non-psychiatrist physicians may prevent far more suicides than further training for psychiatrists.”

Psychiatric illness is present in about 90% of all suicides.

“Suicide risk can be understood in terms of a stress diathesis model (8) (Figure 1) where stress results from an internal stressor, usually a psychiatric illness, present in about 90% of all suicides and most commonly major depression, and/or an external stressor involving life events.”

• cites [8] sourceNot checked — no full text of the source

Acute alcohol or drug abuse adversely influences suicide risk through mood or disinhibition.

“The diathesis moderates suicide risk, and that risk can be influenced by acute alcohol or drug abuse adversely via mood or disinhibition (Figure 1).”

• cites [8] sourceNot checked — no full text of the source

The benefit for suicide in the Garrett Lee Smith evaluation was confined to the demographic group targeted by the intervention.

“These findings indicate the benefit for suicide was found in the intervention counties and confined to the demographic group targeted by the intervention.”

• cites [32] sourceNot checked — no full text of the source

Major depression is the most common psychiatric illness in suicides.

“Suicide risk can be understood in terms of a stress diathesis model (8) (Figure 1) where stress results from an internal stressor, usually a psychiatric illness, present in about 90% of all suicides and most commonly major depression, and/or an external stressor involving life events.”

• cites [8] sourceNot checked — no full text of the source

Combination approaches in healthcare systems show promise for reducing suicide in the United Kingdom, Denmark, and the United States.

“Combination approaches in healthcare systems show promise in reducing suicide in U.K., Denmark and U.S., but evaluating benefit attributable to each component is essential.”

A system-level approach lacks the capacity to measure the separate effect of each intervention component.

“Such an approach lacks the capacity to measure the separate effect of each intervention component.”

A system-level approach combining education, training, and screening shows promise in lowering suicide risk.

“A system-level approach shows promise in lowering suicide risk and involves applying a combination of education, training and screening.”

The 2014 USPSTF report concluded there is insufficient evidence that screening specifically for suicide risk in primary care identifies new cases beyond screening for a psychiatric disorder, distress, or a past suicide attempt.

“The 2014 United States Preventative Services Task Force report concluded there is insufficient evidence that screening, specifically for suicide risk in primary care, identified new cases beyond screening for a psychiatric disorder, distress or a past suicide attempt.(40)”

• cites [40] sourceNot checked — no full text of the source

Only CBT appears to be scalable among psychotherapies.

“Even if psychotherapies were effective, only CBT appears to be scalable (Table 2).”

• cites Table 2Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Education of primary care physicians, and potentially internists and obstetricians, in diagnosis and treatment of depression is a robust suicide prevention approach.

“Education of primary care physicians, and potentially internists and obstetricians, in the diagnosis and treatment of depression is a robust suicide prevention approach.”

Within suicide prevention strategies, there is heterogeneity of study populations in psychiatric illness, proportion of higher risk patients, age, ethnicity, and proportion of males.

“Within strategies there is heterogeneity of study populations in terms of psychiatric illness, proportion of higher risk patients such as those with a history of a past suicide attempt, age, ethnicity and proportion of males (because males have 3–4 times the suicide rate of females).”

The major limitation of this review is the uneven quality and quantity of data available for different suicide prevention strategies.

“The major limitation of this review is the uneven quality and quantity of data available for different suicide prevention strategies.”