Suicide - Washington, 1980-1995

Morbidity and Mortality Weekly Report, June 6, 1997

The ongoing assessment of health data and health data sources is essential to the development of effective prevention strategies for priority health issues. In Washington, assessment efforts include the analysis of suicide data. In 1995, suicide was the eighth leading cause of death in WaShington (1), and most (58%) were firearm related. To determine trends in suicide during 1980-1995, the Injury Prevention Program of the Washington Department of Health (WDOH) analyzed death-certificate data. This report presents the findings of the analysis, which indicate that, while overall suicide rates in Washington remained relatively stable during 1980-1995, suicides became more common among persons aged 15-24 years and [is greater than or equal to] 75 years and less common among persons aged 25-74 years.

Computerized death-certificate data and external cause-of-injury codes (E-codes) were used to identify all suicides (E950-E959) among Washington residents. Population data were derived from the 1980 and 1990 U.S. census and from intercensal and postcensal estimates from the Office of Management of Washington state. Contiguous age categories with similar death rates were grouped, and patterns within age groups were examined.

The average 1-year change in mortality was estimated using negative binomial regression in models that accounted for changes in the age, sex distribution, and size of the population. This regression method is useful for analyzing count data that do not meet the restrictive assumptions of Poisson models (2). Results are expressed as the overall percentage change in mortality from 1980 to 1995. Trends are presented graphically using robust locally weighted regression (3). Because suicide methods might change over time, trends in firearm-related suicides were compared with those in nonfirearm-related suicides.

During 198-1995, a total of 10,650 suicides occurred in Washington, representing an overall average rate of 14.2 per 100,000 population. The most common method of suicide was use of firearms (E950.0-E955.4) (56%), followed by poisoning (E95 E954) (23%), suffocation (E953) (13%), and other or unspecified means (8%). Most (78%) suicides occurred among males. Although the overall average rate of suicide in the total population remained relatively constant during the 16-year period, the rate of firearm-related suicide increased 8% (p=0.2), and the rate of suicide by other means decreased 15% (p [is less than] 0.01) (Table 1). Changes in the overall suicide rate varied by age, increasing by 127% for children aged 5-14 years (all except one suicide in this age group during 1980-1995 occurred among children aged 10-14 years); by 16% for persons aged 15-24 years; and by 42% for persons aged [is greater than or equal to] 75 years (Figure 1). For persons aged 25-74 years, the rate declined substantially. The increase for children aged 5-14 years primarily reflected an increase in nonfirearm-related suicide, the increase for persons aged 15-24 years and [is greater than or equal to] 75 years reflected an increase in firearm-related suicide, and the decrease for persons aged 25-74 years reflected a decrease in both firearm-related and nonfirearm-related suicide (Figure 2).

[Figures 1-2 ILLUSTRATION OMITTED]

TABLE 1. Number and average rate* of suicides, 1980-1995, and overall percentage change in rate from 1980 to 1995, by age group and method of death -- Washington

                             Suicides     Overall change in rate
                                            from 1980 to 1995
Age group (yrs)/                Average
  Method of death         No.    rate       (%)

  5-14
    Firearm                56     0.5     +  69%
    Nonfirearm             47     0.4     + 215%
    Total                 103     0.9     + 127%

  15-24
    Firearm               973     8.6     + 46%
    Nonfirearm            762     6.7     - 14%
    Total               1,735    15.4     + 16%

  25-74
    Firearm             4,200     9.7     - 11%
    Nonfirearm          3,580     8.2     - 19%
    Total               7,780    17.9     - 17%

  [is greater than
    or equal to] 75
    Firearm               705    20.2     + 70%
    Nonfirearm            327     9.4     -  5%
    Total               1,032    29.5     + 42%

  All ages
    Firearm             5,934     7.9     +  8%
    Nonfirearm          4,716     6.3     - 15%
    Total              10,650    14.2     -  3%

                               Overall change in rate
Age group (yrs)/                 from 1980 to 1995
  Method of death      (95% CI ([dagger]))   p value([sections])

  5-14
    Firearm             (-30% to 310%)            0.2
    Nonfirearm           (17% to 753%)            0.02
    Total                (15% to 349%)            0.02

  15-24
    Firearm               (14% to 87%)       [is less than] 0.01
    Nonfirearm           (-34% to 13%)             0.3
    Total                (- 6% to 43%)             0.02

  25-74
    Firearm              (-22% to   2%)            0.09
    Nonfirearm           (-29% to - 6%)      [is less than] 0.01
    Total                (-26% to - 7%)      [is less than] 0.01

  [is greater than
    or equal to] 75
    Firearm              (27% to 128%)      [is less than] 0.001
    Nonfirearm           (-35% to  39%)            0.8
    Total                (10% to  83%)      [is less than] 0.01

  All ages
    Firearm              (- 4% to  21%)            0.2
    Nonfirearm           (-24% to - 4%)      [is less than] 0.01
    Total                (-12% to   7%)            0.6