Understanding Suicide Data

Why is data important to suicide prevention?

Describe patterns: Suicide rates vary across different groups of people, geographic regions, over time, etc. Determining patterns can affect where to put resources for an intervention.

Determine risk factors and protective factors: These are factors that increase (risk factors) or decrease (protective factors)  the chance of a disease or event occurring (e.g. a prior suicide attempt is a risk factor for completing suicide, positive self-esteem is a protective factor).

Project future resource needs: If we know, for example,  that certain age groups are more at risk for suicide then we can target resources to prevention programs for that group.

Suggest hypotheses: Identifying differences in suicide rates across groups often suggests hypotheses for why those differences occur. For example, the western and southeastern states have higher suicide rates, on average. Reasons, or hypotheses, (e.g. differences in racial distribution, gun ownership levels, social isolation) for the regional differences are then proposed and investigated.

Track trends: For example, public health officials may collect data from the Youth Risk Behavior Surveillance System to track changes in the number of high school youth who self-report that they have seriously considered suicide in the past 12 months. If there are increasing reports of suicidal ideation over time, the information may trigger school administrators to implement a prevention program.

Detect epidemics: For example, mortality (death) data may show a larger than expected increase in teen suicides during the last month, indicating that a suicide cluster (a group of suicides that occur closely in time and place) has occurred.

Evaluate prevention programs and policies: If after determining a trend or pattern in suicidal ideation, for example, school officials may decide to implement a peer counseling program.  The school should then track the number of students reporting suicidal ideation on the Youth Risk Behavior Survey or talking to school counselors in order to evaluate whether the peer counseling program is having the desired effect.

How is data gathered?

Mortality data: suicide deaths.

General strengths: Compared with morbidity data, mortality data are generally:

• More completely reported: Legally, all deaths have to be recorded and a death certificate issued. Recording a cause of injury death on a death certificate is legally required.
• More comprehensive: Deaths resulting from suicide are usually investigated and more information is collected than on nonfatal attempts.
• More accessible: Information from all death certificates become part of the national Vital Statistics Records system. This system produces public information that can be easily accessed online and is also used as the basis of many reports and analyses.

General limitations:

• Provide an incomplete picture of the problem of suicidal behavior: Most suicide attempts do not result in death and, by definition, are not included in mortality data.
• Despite better reporting than morbidity data, not all suicides are reported:Sometimes there is not enough information to determine intent. Without conclusive evidence, potential suicides may be recorded as unintentional or undetermined on death certificates. Even if subsequent investigation determines that the death was a suicide, the death certificate may not be updated to reflect this finding. Medical examiners, coroners, doctors, and public safety professionals may not record a death as a suicide to spare the victim and his or her family the social stigma sometimes associated with a death by suicide (or to avoid other consequences such as voiding the victim's life insurance and thereby denying benefit's to the victim's family).
• Inconsistent case definitions about what determines a suicide create difficulty coding mortality data.

Morbidity data: medically-treated, nonfatal suicide attempts.

General strengths:

• Provides a more complete picture: Morbidity data can potentially provide a more complete picture of the problem of suicidal behavior, because most suicide attempts do not result in death.

General limitations:

• Less completely reported: While psychologically serious, many suicide attempts are not medically serious enough to require medical attention and do not get reported/coded. Some attempts that do require medical attention are also not coded as suicide attempts. The social stigma associated with suicide-related behavior can lead victims, families, and medical professionals to conceal self-inflicted injuries. The use of "external cause of injury" codes in hospital data is not required in all states, so attempts may go unrecorded. Inconsistent case definitions for a suicide attempt create difficulty in coding. Stereotypes about who attempts suicide may lead to incorrect diagnoses. 

• Captures a biased view of the suicide injury problem: Hospital datasets are more accessible for public health surveillance than data from private physicians, clinics, and health maintenance organizations. However, hospital data may under- or over-represent certain sub-groups. For example, lower income people are more likely than higher income people to use emergency departments for the care of lower-severity medical problems. As a result, hospital data may over-represent suicide attempts among lower income people. People who are treated for a suicide attempt at a psychiatric facility or a Veteran's Affairs medical facility will also not be captured by hospital discharge data, as these facilities do not participate in the Uniform Hospital Discharge Data System.

Self-reported data: self-reported data on suicidal behavior.

General strengths:

• Provides a more complete picture of individuals suffering from suicidal feelings or behaviors.

• Useful for evaluating trends over time: Because suicide is a relatively rare event, it is difficult measure the impact of a suicide prevention program on mortality data, particularly at the local level. Nonfatal attempts and suicidal ideation are far more frequent and therefore more sensitive to changes over time.

General limitations:

• More difficult to accurately collect data about the way people feel or think versus how they behave.
• Subject to reporting biases. For example, high school students are asked on the Youth Risk Behavior Survey if they ever seriously considered suicide. This question is subject to  recall bias (not all people will remember), social desirability bias (not all will want to admit suicidal feelings, even on an anonymous survey), and to definition issues. After all, what is meant by "seriously" considered suicide?

Source: Information was adapted from the National Center for Suicide Prevention Training online workshop provided by the Suicide Prevention Resource Center, retrieved July 30, 2010.