Roughly 16 percent of people who survive a suicide attempt will make another nonfatal attempt within five years, according to a meta-analysis of 177 studies covering more than 300,000 attempt survivors (Liu et al., Lancet Psychiatry, 2024). That figure is a pooled global estimate, not a personal prediction. The same analysis puts the one-year repetition rate near 8 percent and finds that repetition risk is front-loaded: most repeat attempts cluster in the first 6 to 12 months after discharge from care.
The Mechanism Behind Repetition
Repetition is not simply "the same crisis happening twice." Suicide attempts alter the biology and psychology of the person who survives them. The suicidal state has a cognitive signature: attentional fixation on suicide-related cues, impaired problem-solving under distress, and a narrowed sense of available options. After an attempt, these patterns can consolidate rather than resolve.
Three converging processes drive the five-year rate. The first is acquired capability: repeated exposure to self-injury, including a first attempt, lowers the fear and pain barriers to future attempts. The second is sensitization of the stress response. A first attempt often follows an acute crisis, but the neurobiological aftermath can leave a person more reactive to subsequent stressors.
The third is social and clinical discontinuity. Many attempt survivors are discharged from emergency care into fragmented follow-up. The weeks after discharge carry the highest repetition hazard. Plainly: the attempt itself changes the person's risk profile, and the care system often loses contact exactly when risk peaks.
Quantifying the Five-Year Picture
The 2024 Lancet Psychiatry meta-analysis is the most comprehensive current estimate. Pooled across 177 studies, approximately 16 percent of attempt survivors made a repeat nonfatal attempt within five years. The one-year pooled estimate was about 8 percent, and the two-year estimate about 11 percent. These are aggregate figures drawn from heterogeneous cohorts spanning psychiatric inpatients, emergency department presentations, and community samples.
Absolute counts matter as much as percentages. In the United States, emergency departments treat more than 300,000 self-harm presentations annually (CDC). At a 16 percent five-year repetition rate, that implies tens of thousands of repeat attempts each year traceable to prior presentations. The base rate is large enough that small percentage changes translate into many people.
The rate is also far higher than the base rate of first attempts in the general population. This makes prior attempt status one of the strongest single predictors in suicide risk assessment. A history of attempt is not a "risk factor among many"; it is the dominant clinical signal.
The Timeline of Repeat Attempts
| Time since index attempt | Cumulative repeat attempt estimate | Clinical implication |
|---|---|---|
| 1 month | ~2–3% | Highest per-day hazard; early follow-up critical |
| 3 months | ~4–5% | Crisis plan should be active and tested |
| 6 months | ~6–7% | Post-discharge gap is the danger zone |
| 1 year | ~8% | Most repeaters have already repeated |
| 2 years | ~11% | Slower ongoing accumulation |
| 5 years | ~16% | Long-term monitoring remains relevant |
The shape is steep, then flattening. Roughly half of all five-year repeat attempts occur within the first year. This is why discharge planning, not long-term risk stratification, is where most preventable harm sits. The month after an emergency visit or psychiatric discharge is the single highest-yield window for intervention.
Counterintuitive Finding: Most Attempt Survivors Do Not Repeat
The most important caveat is also the most counterintuitive. About 84 percent of attempt survivors do not make another nonfatal attempt within five years. Even among high-risk subgroups, such as people discharged after psychiatric hospitalization, most do not repeat. This matters clinically and ethically. Treating every prior attempt as near-certain recurrence can produce coercive care, stigma, and lost trust.
A second caveat concerns measurement. Definitional differences change the numbers substantially. Studies counting any self-harm (including non-suicidal self-injury) report higher rates than studies counting suicide attempts with intent to die. Studies using medical records undercount attempts that go untreated. Studies using self-report overcount ambiguous acts. The 16 percent figure is a best pooled approximation, not a fixed constant.
A third caveat is selection bias. Meta-analyses overrepresent people who present to care. People who attempt and never seek treatment are largely invisible in the data. Their repetition rates may be lower, because they may have lower clinical severity, or higher, because they receive no intervention. The honest answer is that we do not know.
Misconceptions About Repetition Risk
Several beliefs about repetition are not supported by the evidence.
Misconception: repetition is inevitable after a first attempt. Reality: most survivors never repeat. Misconception: longer time since attempt means risk is gone. Reality: risk declines but does not reach zero; accumulation continues for years. Misconception: repetition rates are uniform across diagnoses. Reality: rates vary substantially by psychiatric diagnosis, method, and aftercare. Misconception: a "low-lethality" attempt signals low future risk. Reality: intent and medical lethality are imperfectly correlated; method substitution is common. Misconception: hospitalization alone reduces repetition. Reality: discharge into inadequate follow-up can coincide with peak risk.
Practical Actions That Address Repetition
The actionable findings converge on continuity of care. Concrete steps with thresholds and time windows:
Schedule follow-up contact within 7 days of emergency or inpatient discharge; the first month is the highest-hazard window. Use a structured safety plan developed with the patient, not handed to them, and review it at each contact. Prioritize means safety: reducing access to lethal methods during the high-risk window. This is one of the few interventions with consistent evidence for reducing suicide deaths. Maintain contact after missed appointments; loss to follow-up is itself a risk marker. Screen for repetition risk at every mental health contact, and document prior attempts in a way that travels across care settings.
Red-flag warnings for elevated near-term repetition risk:
A first attempt in the past 6 to 12 months, especially with recent discharge. Access to a lethal method combined with acute agitation or hopelessness. Discharge without a scheduled follow-up appointment. A pattern of escalating self-harm or method substitution toward greater lethality.
Limits of the Evidence
The 16 percent estimate rests on pooled observational studies with heterogeneous designs. Significant gaps remain. Estimates come disproportionately from high-income countries with formal mental health systems, so generalization to under-resourced settings is uncertain. Follow-up duration and loss to follow-up vary widely. Few studies distinguish nonfatal repetition from fatal outcomes over the same five-year window.
Risk is also not a single number. It is a distribution across subgroups, and the factors that predict repetition are not identical to those that predict suicide death. Prior attempt predicts both, but repetition is more common than death by a wide margin. Clinicians and researchers should treat the percentage as a population anchor, not a personal forecast. The most defensible use of the number is operational: it shows that repetition is common enough to justify robust follow-up systems and rare enough that most survivors can be helped toward recovery.
This article provides general educational information about suicide-attempt epidemiology. It is not a diagnosis, a treatment plan, or a substitute for individualized clinical assessment. If you or someone you know is in immediate danger, contact local emergency services or a crisis line right away, and consult a qualified local mental health professional for personal guidance.