Which Method of Suicide Is Fatal 90 of the Time: Firearm Injury and the Physiology of Lethality

The method that approaches 90% case fatality is firearm injury, and it does so mainly when the wound reaches the brain, heart, or major central vessels; case fatality is not a fixed property of a method but a product of what tissue is destroyed, how fast blood or airway function is lost, and how quickly care arrives. Published case-fatality estimates for firearm attempts sit near 85–90% in United States hospital and death-certificate analyses, but that figure is an average across wound sites and intent. A shotgun wound to the chest and a grazing wound to the thigh share a category name and almost nothing else. What follows explains the physiology behind lethality, then places every other common method on the same scale.

Mechanism: why firearm wounds kill so efficiently

Lethality in firearm injury comes from three separable mechanisms, and understanding them explains why the 90% figure is both real and conditional. The first is direct tissue destruction along the projectile path. A bullet crushes tissue in a permanent cavity roughly its own diameter and then stretches surrounding tissue outward in a temporary cavity that can be many times the bullet's width. Brain tissue tolerates almost none of this. The cerebral vasculature is fragile, the skull is a closed box, and a wound track through the brainstem or thalamus can abolish breathing and consciousness within seconds. The heart behaves similarly: a hole through a ventricle or the aortic root drops cardiac output toward zero almost immediately, and the pericardium fills with blood that cannot escape fast enough to be compensated.

The second mechanism is hemorrhage rate rather than total blood loss. A person can survive losing a third of blood volume over hours; the same loss over ninety seconds is a different disease. Major arteries and the high-pressure chambers of the heart bleed at rates measured in liters per minute, far faster than the body's compensatory vasoconstriction and tachycardia can offset. This is why anatomic location predicts outcome better than the number of wounds.

The third mechanism is the speed of decompensation. Loss of consciousness from cerebral hypoperfusion can occur within ten to twenty seconds of an arterial injury in the head or chest, and irreversible brain injury begins within four to six minutes without circulation. In plain terms, the clock is shorter than most people assume, and it starts at the moment of injury, not at the moment help is called.

A professional caveat worth stating plainly: case fatality is a ratio of deaths to attempts, so it is shaped as much by who attempts and how they are rescued as by the method itself. The same injury in a rural setting with a thirty-minute transport time and an urban setting with a six-minute one produces different statistics from identical physics.

Quantified context: the numbers behind the ranking

Firearm attempts carry the highest case fatality of any common method in United States data, generally reported in the mid-80s to low-90s percent depending on the dataset and the definition of an attempt. The figure is not a single constant. Studies using hospital discharge and death-certificate linkage tend to report lower values than studies of completed suicides alone, because the denominator includes survived attempts that may never have reached a hospital. That denominator problem is the single largest source of disagreement in the literature.

A useful way to read the numbers is as a gradient rather than a ranking of methods by danger. Firearms sit at the top. Suffocation and hanging follow at roughly half that lethality or less in most United States analyses, though estimates vary widely by how "attempt" is defined. Drowning and jumping from height occupy a middle band that depends heavily on rescue time and height. Poisoning by medication overdose, the most common attempt method by count, is among the least lethal per attempt, largely because the interval between ingestion and death is long enough for intervention.

These ranges are drawn from aggregated United States injury and mortality surveillance and should be read as central tendencies, not guarantees; individual datasets differ by a factor of two in places. What is stable across datasets is the ordering at the top: firearms are consistently the most lethal common method, and overdose is consistently among the least.

Timeline: what happens after a firearm injury

The physiology unfolds on a schedule that explains why the outcome is largely decided before emergency services arrive.

The practical reading of this table is that the decisive interventions are bystander-level: pressure, packing, and tourniquets for extremity bleeding, and CPR if the person is pulseless. Professional care matters enormously, but it arrives after the steepest part of the survival curve has already been traversed.

Edge cases: when the 90% figure does not apply

The 90% figure is a category average, and several conditions move a given injury far from it. Wound site dominates everything. Extremity wounds, even from high-velocity projectiles, are frequently survivable with tourniquet control and modern vascular surgery; the case fatality for isolated limb injury is a small fraction of the overall figure. Abdominal wounds sit in between, with lethality driven by whether major vessels or the liver are involved and by how quickly surgery occurs.

Projectile and weapon type matter as much as anatomy. Shotgun wounds at close range produce massive tissue destruction and carry very high fatality; handgun wounds vary enormously with caliber, bullet design, and range. A through-and-through wound that misses major structures may produce a case fatality closer to that of a serious laceration.

Intent and context are the least discussed and most powerful modifiers. Case-fatality statistics pool impulsive attempts, in which a person may survive because the injury was less destructive or because they or someone nearby sought help immediately, with highly lethal attempts that leave no window for rescue. This is the counterintuitive detail: the same method can appear more or less lethal depending on the population attempting it, which is why method substitution does not reliably change overall suicide mortality in the way a simple lethality ranking would suggest.

Misconceptions: what the "90% fatal" framing gets wrong

The first misconception is that lethality is a property of a method rather than of an injury. Methods are categories; lethality is anatomic and temporal. A firearm is dangerous because of where its projectiles can go and how fast they get there, not because firearms possess an intrinsic fatality percentage.

The second misconception is that a high case fatality means a method is "chosen for" its lethality in every case. Many attempts are impulsive, and the method available in a moment of crisis is often the method used. This matters for prevention: reducing access to highly lethal methods during acute risk periods is one of the few interventions with strong evidence behind it, precisely because the lethal method is often the nearest one.

The third misconception is that surviving an attempt means the danger has passed. Prior attempt is among the strongest known predictors of subsequent death by suicide, and the post-attempt period is a period of elevated risk that warrants immediate clinical engagement.

If you or someone you know is in immediate danger, call or text 988 in the United States to reach the Suicide and Crisis Lifeline, or call 911. These services are free and available 24 hours a day.

Access to a firearm in the home during a crisis period is a major modifiable risk factor. A prior attempt, even a medically minor one, raises subsequent risk substantially. Sudden withdrawal or isolation after a crisis is a warning sign, not a resolution. Giving away possessions or saying goodbye are urgent red flags requiring immediate action. Access to large quantities of medication or to a ligature point are practical risks worth removing.

Practical actions: what the physiology implies you should do

For someone concerned about a person at risk, the evidence-supported steps follow directly from the mechanisms above. Reduce access to the most lethal means first: firearms stored outside the home or locked with ammunition stored separately, medications secured or disposed of, and ligature points made difficult. These are concrete, time-limited actions that match the short window in which impulsive crises occur.

For a person who has already been injured, the priorities are airway, breathing, and hemorrhage control, in that order. Apply direct pressure to bleeding wounds, pack deep wounds with gauze if available, and use a tourniquet on a limb wound that will not stop bleeding. Call emergency services before or while doing this. Do not move a person with a suspected spinal injury unless they are in immediate danger. Do not induce vomiting after an overdose; call poison control at 1-800-222-1222 in the United States.

For clinicians and families managing ongoing risk, the thresholds that matter are behavioral, not numeric: any new access to a lethal method, any statement of intent, and any prior attempt should trigger same-day contact with a mental health professional or crisis service. There is no blood test or score that replaces that judgment.

Limitations: what these numbers cannot tell you

Case-fatality statistics are retrospective and pooled, so they cannot predict an individual outcome, and they should never be read as a statement that any method is survivable or unsurvivable. Definitions of "attempt" vary between datasets, which is why the same method can appear with a fatality rate differing by twenty points across studies. Much of the underlying surveillance is United States-specific and may not generalize to settings with different firearm access, emergency response times, or trauma systems. The mechanisms described here are stable physiology; the percentages are population estimates with wide confidence intervals, and the ranges in the table above are presented as central tendencies rather than precise values. Where the evidence does not support a specific figure, this article states the mechanism and the direction of effect rather than inventing a number.

This article is general educational information about injury physiology and population statistics, not a diagnosis, treatment plan, or clinical recommendation for any individual. It does not promise any outcome and should not be used to assess a specific person's risk or prognosis. If you or someone else may be in danger, contact local emergency services or a crisis line immediately, and consult a qualified local clinician for personal medical guidance.