J. Biomedical Science and Engineering, 2009, 2, 516-520
doi: 10.4236/jbise.2009.27074 Published Online November 2009 (http://www.SciRP.org/journal/jbise/ JBiSE
).
Published Online November 2009 in SciRes. http://www.scirp.org/journal/jbise
Efficiency of diagnostic model to predict recurrent suicidal
incidents in diverse world communities
Vatsalya Vatsalya1, Kan Chandras2, Shweta Sri va st a va3, Robert Karch1
1Health Promotion Management American University, Washington D.C., USA; 2Behavioral Sciences, Fort Valley State University,
Fort valley GA, USA; 3Department of Biology, Georgetown University, Washington D.C., USA.
Email: vvatsalya@gmail.com
Received 16 July 2009; revised 26 August 2009; accepted 27 August 2009.
ABSTRACT
Suicidal attempts have a very significant effect on the
society, and they also reflect on the efforts of the
supporting health care and counseling facilities; and
the mental health professionals involved. The impact
of suicide is further magnified by the needs of per-
sons who attempt suicide multiple times, requiring
emergency health care and rehabilitation. Preve nt ing
such activ ities becomes a ma jor task for the support
providing agencies as soon as patient with such ten-
dencies are identified. There are repetitive traits
that can be observed during the entire therapeutic
program among the high-risk group individuals,
who are susceptible to this kind of activity and such
traits indicate for specific profiling. The aim of the
instrument is to prevent the occurrence of the re-
petitive suicidal attempts of the patients in various
world regions, which may have significantly higher
and concerning suicide rates. This profile has been
constructed on the various parameters recognized in
the statistical analysis of the patient population,
which have been identified or can be under treat-
ment for their suicidal behavior. This instrument is
developed to predict the probability of population
segments who may attempt suicide and repetitively,
by matching the parameters of the profile with that
of the patient pool. Building a profile for the purpose
of predicting behavior of this kind can strengthen the
intervention strategies more comprehensively and
reduce such incidents and health care requirements
and expenses.
Keywords: Instrument; Parameters; Predictor; Risk Cate-
gory; Suicidse
1. INTRODUCTION
Several world regions face the emerging concern of sui-
cidal incidents and such growth patterns have suggested
typical geographical factors playing major contributory
roles [1]. Many regions so far have been identified with
consistent year wise rising suicide rate and from such
locations, Washington State has been included in the
study to evaluate this phenomenon and derive a model to
investigate and predict suicidal incidents in the general
populations. Nationally, Washington State has the 16th
highest suicide rate and suicide is the eleventh leading
cause of death with 814 death reported in year 2005 with
a rate of 13.1 incidents per 100,000 (age adjusted) of
population, compared to the national US rate of ap-
proximately 11.0 [2]. Suicidal attempts can be five time
or more than the suicide death rate [3] (Figure 1).
There have been consistencies in reports that even af-
ter getting initial medical and other support services,
patients attempt suicide again in Washington State with
similar rate as without the interventions [4]. During
2003–2005, males in Washington accounted for 79% of
completed suicides, with the highest rate of suicide, observed
in the group among the age group of 75 years and more
[5]. In 2004; in the 15–24 age groups, suicide was the
second most leading cause of death and altogether 17%
deaths [6]. Nearly two-thirds of teens with clinical de-
pression go unnoticed and may not get treated; among
these, males 15 to 19 years old are five times more likely
Figure 1. Yearly suicide rate representation per 100,000 popu-
lations (Suicide attempt and death rate versus years as x:y co-
ordinates).
V. Vatsalya et al. / J. Biomedical Science and Engineering 2 (2009) 516-520 517
Figure 2. Age-group based comparisons of number of suicidal
incidents (Age-groups versus number of incidents as x:y coor-
dinates).
than females to complete suicides [7] (Figure 2). The
frequency of suicide attempts can increase, if the causa-
tive factors are not resolved. Extent for emotional and
physical pain cannot be assessed for individuals and
their relatives, though approximately three billion US
dollars of loss associated with the suicidal incidents has
been recorded so far during the year 2002–2006 [8].
Preventive procedures have been particularly promoted
to reduce such incidences, which have priority concern
to precisely identify target population. Development of a
predicting instrument, which can provide appropriate
behavioral evaluation, can be one of the choices to this
requirement.
This instrument is based on set of primary repetitive
characters, which can provide a predictable analysis of
such behavior among the high-risk group individuals,
who are susceptible to commit or attempt suicide. The
concept of predictor is to identify the patients with such
tendency or ideations and inform the healthcare profes-
sionals and people concerned, to prepare and implement
specific therapeutic plan to minimize such incidents fur-
ther. A predictor profile for the purpose of interpreting
specific behavior would strengthen the intervention
strategies, reduce the frequency of such incidents, and
regularize involvement of health care and financial re-
sources.
2. METHODOLOGY
Data source for the profile analysis has been collected
from yearly published hospitalization reports of suicide
incidents from Washington State public health statistical
data bank, Washington State Injury and Violence Pre-
vention Program, 2008; and GIS from 2002 to 2006 year.
Data from 15,826 cases with suicides and single or mul-
tiple identified attempts of history during the interval
2002–2006, betw een the age group of 15 and 75 + years,
have been included in the study (Figure 3). Significant
cohort parameters have been studied from the target
population for the development of the instrument,
namely gender, age groups, health condition, biological
and environmental factors (namely economic, academic,
social, ethnic and likewise). Statistical analysis has been
performed to identify repetitive characteristics of the
parameters with annual rate and frequency comparisons.
An analog model is constructed utilizing the parameters
and their characteristics under observations. The charac-
teristics have been broadly grouped as parameters of
interests and highly repetitive prevalent characteristics
observed in the evaluation have been consigned under
high risk category (Colored in red, Figure 6); average
repeating characteristics as intermediate and less fre-
quently occurring as low risk category. These parameters
were constructed in the form a pathway with the promi-
Figure 3. Methodology illustration for investigating model parameters.
SciRes
Copyright © 2009 JBiSE
V. Vatsalya et al. / J. Biomedical Science and Engineering 2 (2009) 516-520
518
nent parameters evaluated initially within which all the
characteristics of each parameter undergo comparison
for in track corresponding profile similarities.
Characteristics which have not demonstrated signifi-
cant prevalence in the analysis or are not significantly
identifiable in the data search and evaluations have not
been included in the development of the analog at this
stage of the study. A follow up evaluation assessment has
been conducted for comparing this profile with the data
from new reports to evaluate the predictability th e evalu-
ations and to develop the validity of the model.
3. RESULTS
The parameters are assessed in relation to the numbers
of attempts and actual suicide incidents reported be-
tween year 2002 and 2006. The age groups 15–24 and
35–44 yrs have the maximum incidents of attempted
suicides (Figure 2).
The leading conversion of the attempts into actual su i-
cide is demonstrated by the age group 45–54 yrs and
from 2002 year onwards, there is a steady increase in the
rate of suicide and suicide attempts in this age group
(Figure 4). Males in Washington accounted for 79% of
total suicides, though majority of females have at-
tempted suicides with a rate of 62%. There has been a
consistent annual rise in the number of su icidal attempts
as evident from annual health reports (Figure 5). Suicide
comparison ratio among the adult male and female pa-
tient population rate has been measured as 7:2, which is
evident from the finding that during 2003–2005 period.
Among adolescen t and young age group s, approximatel y
25% of attempted suicide by youth male results in death;
whereas three–four percent deaths result in females. 15
to 19 year old males are five times more likely than fe-
males to commit suicides. Caucasian and American In-
dians demonstrate largest proportion of actual number of
suicidal attempts and incidents as 14 every 100,000
resident population each.
Data reflects that in Caucasian and Native Indian
groups, suicide is occurring with 14 per 100,000 rates
each though African-American, Asian and Hispanic
groups demonstrated non-significant rate of eight, eight
and six consecutively. High competition, economic and
social hardship can be observable factors among Cauca-
sian though relationship cannot be established directly;
partially available particulars are only confirmed presently
for family predisposition, abuse, addiction and health
concerns among with Native Americans too [9]. Suicide
rate as observed in relation with academic background
can be described with the ratio of 28:14:9 with 28 with
patients having 12th grade or less education, 14 with
some college level programs and 9 with graduation or
post graduation. Linear regression of suicide rate has
been observed with progression of academics in the sta-
tistical evaluation. Poisoning is identified as the most
common means for attempted suicide cases among
15–24 and 35–54 age group whereas firearms are as the
most common means among the young and adu lt groups
for suicide deaths. Based on the results of statistical
analysis and evaluations, several parameters and their
characteristics have been identified and incorporated for
the development of th e mode l fo r the instrumen t with th e
order as gender, age group, ethnicity, health, education,
environmental condition, and; exposure and access to
perform self-fatality for the Washington region (Fi gure 6).
The construction of the analog utilizes major and re-
curring parameters first and tracks subsequent parame-
ters and their characteristics in the descending order of
evaluation. These Follow-up evaluations with parame-
ters evaluated from various new case studies (a total of
627 case data) rando mly, in dicated results in compliance
with an accuracy of more than 90% of the total cases,
when the parameters are run through the analog model.
4. DISCUSSION
Age groups 15–24 and 35–54 year have demonstrated
the majority of incidents of suicide and su icidal attempts.
Parallel emergence of economic hardship can be pro-
jected as one of the reasons for the growth of suicide rate
among the 45–54 year age group [10].
Higher suicide rate among is recorded among males,
approximately 25% of attempted suicide by youth male
results in death, whereas 3–4 % death results in females,
though higher attempt rate among females is observed
[11]. Health conditions (depression, substance abuse,
Figure 4. Yearly distribution of suicidal incidents in various
age-groups (Number of incidents versus years as x:y coordi-
nates).
Figure 5. Yearly distribution of total number of suicidal inci-
dents (Years versus number of incidents as x:y coordinates).
811 801 823 814 796
2901 3007 3209 3507 3445
0
500
1000
1500
2000
2500
3000
3500
4000
'02 '03 '04 '05 '06
Suicide Death
Frequency
Suicide Attem pt
Frequency
Linear (Suicide
Attempt Frequency )
716
717
818
840
810
680
641
727
767
742
841
770
817
804
826
490
553
579
626
677
157
162
193
193
244
48
69
40
59
40
58
48
43
60
0200 400 600 800 1000
'02
'04
71
'06
75 yrs +
'05 6
5-7
4
5
5-6
4
4
5-5
4
3
5-4
4
2
'03 5-3
4
1
5-2
4
SciRes
Copyright © 2009 JBiSE
V. Vatsalya et al. / J. Biomedical Science and Engineering 2 (2009) 516-520 519
Figure 6. Analog construction of diagnostic model.
ideation and cond itions like family history, sexual orien-
tation and abuse, disability) show higher probability of
suicidal incidents (Figure 7). Individuals with physical
and sexual abuse have been observed to have four times
higher suicide attempt rate than non-abused. Among
African American communities, some of the major fac-
tors are community segregation, economic hardship and
access to weapon; though cultural encapsulation and
stress is prevalent among other minority groups. From
2002 year onwards, there is a steady increase yearly, in
the number and rate of suicide incidents in the 35–54 age
groups [12]; and the proportion of people living in pov-
erty increased, though data is not sufficient enough to
identify relation to economic hardship [13]. Post-Trau-
matic Stress Disorder, economic hardship, social con-
structs are being studied as an extension of the ongoing
study.
546
600 Male Suicide
539
483
500 459 Frequency
478
Male suide ic
Attempt
400 327
300 294
279
5. CONCLUSIONS
The complexity of suicidal behavior and ideation re-
quires multiple and immediate preventive approach and
application of a predictor model can supplement target-
ing population, requiring adequate health care services to
prevent suicide. Since much published material and
clinical experience demonstrate a number of causative
factors associated with suicide, the early identification
and appropriate treatment of this condition is an impor-
tant strategy for prevention. Improvements may be re-
quired for broadening data collection and investigatory
techniques for possible suicidal attempts, social con-
structs and economic causes; biological and environ-
mental factors and minority groups from both fatal and
Figure 7. Yearly distribution of depression-originated youth
suicidal incidents (Years versus number of incidents as x:y
coordinates).
non-fatal suicide incidents. Utilization of the instrument
at the key regions and community level; significantly
social and community based agencies, academic institu-
tions and care centers can be highly valuable and rec-
ommended. Design, planning and Implementation of
intervention strategies for newly growing suicidal inci-
dents in the 45–54 year age group should be focused
appropriately. A preliminary investigation in various
world regions with above-average suicide rate have
identified points of interests namely Kutznetsk Basin,
former USSR republics (between 1980 and 1995 citing
economic instability); among indigenous peoples in bo th
Australia and Canada in the last 20 years [1]. Depending
upon the various factors and parameters in different
world locations, with specific standardization, an appro-
priate constru ction of the predictor mod el and evalua tion
may be utilized for similar diagnostic procedures and
further research may provide essentials and scope for
development for area specific revisions.
78 82 78 85 99
238 247 Female Suicide
200 Frequency
Female Suicide
100
16
14 Attemp
t
24
18
16
0Linear (Male Suicide
Frequency)
2002 2003200420052006 Year
SciRes
Copyright © 2009 JBiSE
V. Vatsalya et al. / J. Biomedical Science and Engineering 2 (2009) 516-520
520
REFERENCES
[1] World Report on Violence and Health: Self-Directed
Violence (1999). Geneva, Wold Health Organization
(document WHO/ HSC/PVI/99.11).
[2] Injury and Violence Prevention Program. Washington
State Department of Health [online], 2007, http://www.
doh.wa.gov/hsqa/emstrauma/injury/data-tables.
[3] Washington State Department of Health, (2004). Suicide
chapter, The Health of Washington State 2004 Supple-
ment. Olympia, WA, http://www.doh.wa.gov/HWS/HWS
2004supp.htm.
[4] American Journal of Psychiatry [online], 2006,
http://ajp.psyh.iatryonline.org/cgi.
[5] Health of Washington State Report. Washington State
Department of Health [online], 2007,
http://www.doh.wa.gov/HWS/default.htm.
[6] Simon, G. E., Savarino, J., Operskalski, B., and Wang, P.
S., (2006) Suicide risk during antidepressant treatment,
Am. J. Psychiatry, 163, 41–47
[7] Center of Health Statistics (Death data). Washington
State Department of Health [online], (2007), http://www.
doh.wa.gov/ehsphl/chs-data/death/dea_VD.htm.
[8] Center of Health Statistics (Hospital data). Washington
State Department of Health [online], (2007), http://www.
doh.wa.gov/EHSPHL/hospdata/default.htm.
[9] Mayer-Gross, W., Slater, E., and Roth, M., (1960) Clini-
cal Psychiatry, London, Cassell.
[10] U.S. Centers for Disease Control and Prevention, Na-
tional Centers for Injury Prevention and Control. (2007).
Web-based Injury Statistics Query and Reporting System
(WISQARS), Retrieved January 3, 2007 from http://www.
cdc.gov/ncipc/wisqars.
[11] Rotheram-Borus, M. J., Piacentini, J., Cantwell, C.,
Beline, T. R., and Sone, J. (2000). The 18-month impact
of an emergency room intervention for adolescent female
suicide attempters. Journal of Counseling and Clinical
Psychology, 68(6), 1081–1093.
[12] Centers for Disease Control and Prevention (CDC).
Web-based Injury Statistics Query and Reporting System
(WISQARS) [Online], (2005) National Center for Injury
Prevention and Control, CDC (producer), Available from
http://www.cdc.gov/ncipc/wisqars/default.htm.
[13] Geographic Information Systems. Division of Informa-
tion Resource Management, Washington State Depart-
ment of Health [online], 2007,
http://www.doh.wa.gov/gis/gisdata.htm.
SciRes
Copyright © 2009 JBiSE