causal relationships for suicide is not possible. (3) Risk factors can change in individuals over short periods of time, and across developmental life-stages, further complicating assessment of suicide risk and prevention.

Prevention Framework. Dr. Brown described three levels of the currently most broadly used conceptualization of prevention. (1) Universal interventions are delivered to everyone, regardless of an individual’s vulnerability. The law requiring seat belt use is an example. (2) Selected interventions are delivered to groups at increased risk. Anti-alcohol abuse classes for adolescents is an example. (3) Indicated prevention is given to individuals who have a known risk factor. For suicide prevention, treating a person with depression is an indicated prevention.

If you are doing a selective or indicated intervention, the only impact you are going to have is on those people who you have identified. You have no impact on those who are at risk, but you missed.

C.Hendricks Brown

According to Dr. Brown the best prevention programs combine interventions on more than one level. Such “unified intervention” strategies are more effective and more practical. They avoid problems faced when individuals’ risk factors change over time. Programs using only one level of intervention have to exclude individuals whose level of risk changes during the program, possibly putting these people at risk if exclusion from the program means they lost their only available treatment. In addition, information about the common occurrence of changing risk levels is lost, hindering evaluation of the program.

Prevention Program Strategies. Dr. Brown listed four general strategies for prevention. They are: developmental approaches, individual approaches (reducing risks and increasing protective factors), intervention through policy and law, and intervention through the social context (e.g., reducing poverty, child abuse). Dr. Brown described a promising new strategy of the last type, called “The Empowerment Intervention.” In this design, two communities are randomly assigned to either starting a prevention program immediately, or waiting 1 year before starting. Five community pairs make up the data set. This allows collection of “good quantitative data that is not confounded with community readiness and other community characteristics,” according to Dr. Brown.

Dr. Brown noted that it is important to take into account two proportions when trying to estimate the number of people targeted by a particular prevention effort. The first is “population-attributable risk.” This is the proportionate reduction in number of cases of the condition that would occur upon elimination of a particular risk factor from the population at large. For example, one risk factor associated with the vast majority of completed suicides is depression, according to Dr. Brown. Therefore, if depression could be eradicated from the general population, the vast majority of suicides, according to Dr. Brown’s explanation, would be prevented. The second proportion to consider when designing prevention efforts is the “relative risk.” The rela-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement