Roles of physical and mental health in suicidal ideation in Canadian armed forces regular force veterans.
Thompson, James M. ; Zamorski, Mark A. ; Sweet, Jill 等
Suicide is an important public health problem and therefore is of
interest
to Veterans Affairs Canada (VAC) and similar agencies worldwide that
support the well-being of veterans and their families. The protracted
conflicts in Afghanistan and Iraq have triggered renewed interest in the
life-course health effects of military service, including suicide. (1,2)
Historically, suicide rates among serving and released (veteran)
military populations have tended to be similar or lower than in civilian
populations, consistent with a healthy worker effect. (2-5) Suicide
rates have increased in the US and UK in the past decade among serving
personnel (2-4,6-9) and in some veteran subpopulations, including young
males and those with chronic health problems. (3,6-8,10,11) A mortality
study of Canadian Armed Forces personnel who enrolled in 1972-2005 found
elevated rates among veterans. (4)
Suicide risk in contemporary veterans needs explanation. (2) Mental
health problems are strongly associated with suicidality (ideation,
attempt and death) and are known military occupational risks.
(1,2,11-19) In an expert consensus framework proposed for suicide
prevention, stressful life events, psychiatric disorders and
dispositional risks, including personality and social factors, were
viewed as key pathways toward suicidal behaviour, but physical health
was not included. (20) The possibility that physical health also plays a
role in explaining suicide risk has not been well explored in veterans.
A broad range of chronic physical health conditions also occur in
military service. (1) In veterans, as in the general public, physical
conditions are about 2-4 times more prevalent than diagnosed mental
health conditions, and the prevalence of some physical conditions,
particularly musculoskeletal disorders, has been higher in some veteran
populations than among comparable civilians. (21,22) Age and sex have
been consistently associated with suicidality, and socio-economic
factors, including marital status, education and financial status, have
been associated with suicidality as influencing factors.
(2,4,5,7,10-13,23-30) Military rank has been found to be associated with
Canadian and UK veteran suicide. (4,8)
Associations between physical health and suicidality are supported
by a developing body of evidence in civilian studies, particularly of
chronic pain, (10,23-26,31) but are less well established than for
mental disorders owing to inconsistent findings and methodological
variability. (25) Physical health has been little studied as a risk
factor for veteran suicidal ideation. In US veteran studies, severe pain
was associated with suicide after adjustment for psychiatric conditions
and other factors; (31) life stress related to physical health in the
two weeks before suicide was associated with completed suicide; (12) and
physical health difficulties were associated with suicidal ideation in
older veterans. (27)
The current study is unique in exploring associations between
past-year suicidal ideation and physical and mental health from a survey
of recent canadian veterans. Although most of those with suicidal
ideation do not progress to suicide, ideation is the entry to suicidal
pathways, one of the strongest predictors of suicide and a key
indication for intervention to prevent suicide. (2,18) The objectives of
this study were to examine whether impaired physical health was
associated with suicidal ideation; whether such a relationship was
explained by the association of physical and mental health; and whether
mental and physical health interacted to increase risk for ideation.
METHODS
The 2010 Survey on Transition to Civilian Life (STCL) involved
canadian Armed Forces Regular Force veterans who had been released from
service during 1998-2007. (21) They enrolled in service from the 1960s
to the 2000s and had varied experience in training, domestic disaster
response and international peacekeeping, primarily in Cyprus and the
Balkans, the first Persian Gulf War in 1990-91 and, more recently, in
the combat and peace support mission in Afghanistan.
Sample
The cross-sectional, computer-assisted telephone interview survey
sampled former Regular Force personnel who had been released from
service during January 1998 to December 2007, had not reenrolled and
were not living in institutions, the northern Territories or outside of
Canada. (21) The sample was established by Statistics Canada in November
2009 using the Department of National Defence (DND) human resources
database. Contact information was obtained by Statistics Canada through
linkage with administrative data. Statistics Canada's Policy
Committee provided approval for the study. Of the 4,721 former Regular
Force personnel sampled, 3,355 responded (response rate 71%). Of these,
3,154 (94%) agreed to share their data with VAC and DND. After variables
had been selected for this analysis, complete records were available for
2,658. There were no significant differences in the prevalence rates of
ideation, socio-demographic characteristics or health conditions between
the full STCL sample and the analyzed subsample.
Suicidal ideation
Ideation was assessed using questions derived from the Canadian
Community Health Survey. All respondents were asked, "Have you ever
seriously considered committing suicide or taking your own life?"
Those who said "yes" were asked, "Has this happened in
the past 12 months?"
Physical and mental health covariates
Three health measures were used in the study. First, respondents
were asked whether they had chronic health conditions diagnosed by a
health professional that had lasted or were expected to last six months
or more. Physical health conditions were grouped: musculoskeletal
conditions (arthritis or back problems excluding fibromyalgia);
cardiovascular conditions (high blood pressure, heart condition, effects
of stroke); respiratory conditions (asthma, emphysema, chronic
bronchitis and chronic obstructive pulmonary disease); gastrointestinal
conditions (intestinal or stomach ulcers, bowel disorder such as
Crohn's disease, ulcerative colitis, irritable bowel syndrome and
bowel incontinence); hearing problems; chronic pain or discomfort;
obesity; and diabetes. Cancer was excluded since fewer than five
veterans with cancer had suicidal ideation. Mental health conditions
were assessed using four categories: mood disorder, anxiety disorder,
anxiety or depression, and post-traumatic stress disorder. The second
health measure was the number of conditions. The third was Version 1.0
of QualityMetric's SF-12 Health Survey, which measures both the
presence of health problems and health-related functioning. Physical and
mental component summaries for individuals were computed using
QualityMetric's software; lower scores reflect poorer health.
Socio-demographic and military covariates
Following exploration of ideation prevalence (28) and examination
of the literature, we derived an initial list of 10 potential
covariates: age, sex, marital status, education, income, rank, service
branch, length of service, enrolment era and deployment outside Canada
for 30 days or more for combat, peace-keeping, humanitarian aid or
non-routine deployments to sea, excluding training. Age at survey, sex
and rank at release were obtained from the DND database. Ranks included
categories for commissioned officers (cadet and officer ranks) and
non-commissioned members (recruit, junior and senior ranks). Other
variables were self-reported. Educational level attained was grouped
into four categories: less than high school graduation, high school
graduate, some postsecondary education and university graduate.
Statistics Canada's Low Income Measure is based on household income
and number of people in the household.21 Correlations with ideation were
assessed using SPSS (Version 16.0, SPSS Inc., Chicago, IL) statistical
tests for categorical and continuous variables and unadjusted binary
logistic regressions in Stata version 11.1 (StataCorp, College Station,
TX, 2011); those with significant associations were included in the
regression models.
Statistical analysis
Analyses were conducted for respondents with complete ideation
data. Stata was used to account for sampling stratification by VAC
benefit recipiency status, age, sex and non-response.21 Taylor series
linearization was used to calculate confidence intervals. We initially
explored associations of ideation with the socio-demographic covariates
(age, sex, marital status, income, education and rank) and the health
measures using unadjusted odds ratios (UORs) in separate logistic
regressions. Separate multivariate regressions were conducted with each
of the three health measures owing to collinearity concerns, with
adjustment for the six sociodemographic variables in each case (Table
1).
RESULTS
Sample characteristics
The mean number of years between release from service and
participation in the survey was 6.8 (standard deviation [SD] 3.0, range
2.3-12.2). Mean age was 43.5 years (SD 11.2, range 20-67). Few had low
household income (7%, similar to the 2009 age- and sex-adjusted general
Canadian population), (21) 11% were women, 53% had at least some
post-secondary education, and 75% were in married or common-law
relationships (Table 1). Most (64%) were released with non-commissioned
ranks, 60% had been deployed outside Canada for 30 days or more, and 25%
were released owing to medical employability limitations. Nearly all of
the 34% who were participating in VAC programs were eligible for
disability benefits because of service-related physical and/or mental
health diagnoses by administrative criteria. There was a substantial
prevalence of self-reported diagnosed chronic health conditions: 81% had
one or more diagnosed physical conditions, and 23% had one or more
diagnosed mental conditions. Almost all with mental conditions had
co-morbid physical conditions (95%), whereas only about a third with
physical conditions had mental conditions (27%). The weighted population
prevalence of suicidal ideation was 5.8% (95% CI: 5.0%-6.8%) and of
past-year suicide attempts was 1.1% (95% CI: 0.7%-1.6%).
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Unadjusted associations with ideation
Significantly elevated UORs were found for female sex, middle age
(35-49 years), widowed/separated/divorced, lower educational attainment,
low income and non-commissioned rank at release (Table 1). UORs were
significant for all groups of health conditions but were much lower for
physical (1.7-3.6) than mental conditions (7.0-11.6) (Table 1). Age was
used in the multivariate models instead of two variables strongly
correlated with age: length of service (Pearson's coefficient 0.90,
p < 0.01) and enrolment era (Kendall's tau-b, -0.80, p <
0.01).
Figure 1 shows that increasing numbers of physical and mental
conditions were associated with ideation in approximately linear
relationships. Figure 2 shows that decreasing SF-12 summary scores
(worse health) were associated with higher ideation in an approximately
linear fashion.
Those with three or more physical conditions and at least one
mental condition had a high prevalence of ideation (21.5%, CI:
18.0%-25.5%) and markedly increased odds of ideation (UOR 17.3, CI:
9.5-31.6) relative to those with two or fewer physical and no mental
conditions. Moreover, this high co-morbidity group contained only 15.2%
(CI: 14.0%-16.5%) of the overall STCL population but contained 55.9%
(CI: 47.7%-63.8%) of all ideation cases (Figure 3).
Associations with ideation after adjustment
All three health measures were independently associated with
ideation after adjustment for the socio-demographic characteristics
(Table 1). After adjustment for socio-demographics and the presence of
health conditions in a single multivariate model there was an
independent association with ideation only for gastrointestinal
disorders (adjusted odds ratio [AOR] 1.66, CI: 1.032.65); stronger,
persistent associations among the mental conditions were found only for
the "depression or anxiety" category (AOR 5.06, CI: 2.97-8.62)
and mood disorders (AOR 2.91, CI: 1.67-5.07). In the number of
conditions model, the AOR of ideation for each additional mental
condition was 2.32 (CI: 2.012.68) and for each additional physical
condition was smaller, at 1.22 (CI: 1.05-1.42). In the SF-12 model,
current mental health status was more strongly associated with ideation
than physical health: each one-point increase in the mental summary
decreased the odds of ideation by 0.88 (CI: 0.87-0.89), while each
one-point increase in the physical summary decreased the odds by 0.98
(CI: 0.96-0.99). Of the socio-demographic characteristics, only marital
status other than married or common law remained significantly
associated with ideation.
Multiplicative interaction terms using products of number of mental
and physical conditions and SF-12 scores were not significant in
multivariate models (not shown) so were not included in the final
models. Figure 4 suggests the possibility of an additive, synergistic
interaction between physical and mental health, as there was a stronger
association of ideation prevalence with increasing physical health
co-morbidity in the presence of at least one mental condition.
DISCUSSION
This study found that mental health had a strong, independent
association with suicidal ideation and that physical health also was
independently, though less strongly, associated with ideation. Ideation
prevalence was highest among veterans with multiple physical conditions
and co-occurrence of one or more mental conditions, and ideation cases
were heavily concentrated in this highly co-morbid group. There was a
suggestion of a possible synergistic, additive interaction between
physical and mental health in association with ideation. In this study,
5.8% of Regular Force veterans had ideation compared with the 4.0%
prevalence found nearly a decade earlier in 2002 for both the Canadian
general population and serving personnel. (5) These prevalence rates are
not directly comparable because of differences in era, age, sex
composition and survey methodology.
The strong association of mental health with ideation found in this
study is consistent with well-established evidence from prior research
in civilian and military populations. The finding of independent
associations of both physical and mental health with ideation lends
support to the hypothesis that the known effect of military service on
health (1,2,17) could be one mechanism contributing to increased suicide
in some veteran populations.
For all health measures used in this study, controlling for mental
health and socio-demographic characteristics attenuated the unadjusted
association of physical health with ideation. This finding accords with
previous studies in general populations, although associations with
specific conditions have been inconsistent. (23,24) Physical health was
found to be one of the strongest risk factors for suicidal ideation in
older US veterans. (27) The finding that the unadjusted associations
between physical health and ideation were attenuated by adjustment for
mental health is consistent with the explanation that much of the effect
of physical health on ideation is mediated by mental health, which is
plausible and consistent with prior work. (14,25) Such mediation is also
consistent with theoretical frameworks proposed for suicidality in
military personnel and veterans, which emphasize the importance of
psychological and coping factors arising from individual predisposition
and perceived social distress. (2,7,15,24-27,29) Mediation is also
consistent with the Mann suicide prevention framework, (20) in that
physical health problems could be antecedents to mental disorders or
stressful life events that drive suicidality. (29) Our findings suggest
paths between physical health and ideation that are not reflected in the
Mann framework and remain to be clarified.
This study found no direct association between suicidal ideation
and deployment, a finding consistent with prior studies that have
controlled for military and other factors. (7,15,16,18,30-32) The STCL
did not assess deployment characteristics (such as combat exposure)
known to be associated with mental health problems; however, we
controlled for mental disorders. Although combat exposure can result in
physical and mental health problems for some, nonmilitary factors have
been associated with suicidality in veterans. These include protective
factors, such as resilience (14) and social support, (9,16,27) and
predisposing factors, such as social readjustment stressors, (26) low
social support, (9) age, male sex, White race and preservice factors.
(7,26,30) The only socio-demographic variable that remained significant
in the multivariate models was marital status other than married or
common law, likely signalling the importance of social support. This is
consistent with emerging evidence that military population suicides are
associated with a range of factors similar to those experienced by
civilians. (2,5,7,9,10,12,14,18,26,30) It seems likely that the
interplay of later life health status and individuals'
predispositions and post-service psychosocial factors could account for
increased suicidal ideation in some veteran populations. (9) Additional
research is needed to explore the relationships of physical health with
these potential mediating factors.
Strengths and limitations
This was the first comprehensive, national study of suicidal
ideation in recent veterans living in the general Canadian population.
The response rate was good, and the sample was statistically
representative of Regular Force personnel who were released during
1998-2007. Sampling weights provided by Statistics Canada accounted for
age and sex differences between responders and non-responders. Veteran
identity and some socio-demographic and military characteristics were
objectively determined through data linkage.
The study was cross-sectional, so associations cannot be presumed
to imply causality. Use of self-reported diagnosed health condition
groups did not include undiagnosed problems, and specific diagnoses
within groups of conditions vary in terms of severity, natural history
and predisposition to suicidality. However, the similarity in the main
findings using all three ways of measuring health status is reassuring,
since SF-12 scores account for the functional impact of undiagnosed as
well as diagnosed conditions. The mental condition questions did not
identify mutually exclusive groups; however, there was a more or less
linear association between number of mental conditions and ideation.
Interactions can be difficult to detect using logistic regression for a
low prevalence outcome such as suicidal ideation, and very few veterans
had mental conditions without physical conditions. The sample was too
small to investigate the role of physical health in suicide attempts,
and it is conceivable that physical health could play a different role
in life course pathways from ideation to suicidal behaviour. The survey
did not sample all Regular Force veterans living in Canada and included
no Reserve Force veterans, so while the findings are not representative
of the much larger veteran population, they are representative of
modern-day Regular Force veterans exposed to the increased operational
tempo of Canadian Armed Forces operations since 1990.
CONCLUSIONS
These findings suggest that the current singular focus on
psychosocial problems in contemporary veterans should not eclipse the
potential impact of physical health problems. Attention to physical
health is recommended when diagnosing and treating mental health and
social problems. (6) The findings support policies that provide access
to mitigation of both physical and mental health problems. Subgroups
with a higher prevalence of suicidal ideation could be targets for
resource allocation in suicide prevention, particularly veterans with
multiple chronic physical conditions and co-occurrence of physical and
mental conditions. Lack of an association of ideation with deployment
suggests that suicide might be approached as a public health problem in
the broader veteran community as opposed to a deployment health problem
in war veterans.
Physical health is a potentially important factor in research
determining pathways to suicidal ideation in veterans and potentially to
suicidal behaviour. Larger samples, prospective studies, more precise
measurement of health conditions using structured symptom measures,
clinical records and measures of health condition severity would clarify
the smaller direct effect of physical health and its possible
interactive effect with mental health. These approaches would clarify
whether the small, independent association between physical health and
ideation is due to a strong effect in a small subpopulation of veterans
or a weak effect in many.
Received: August 20, 2013
Accepted: February 12, 2014
REFERENCES
(1.) Levy BS, Sidel VW. Health effects of combat: A life-course
perspective. Annu Rev Public Health 2009; 30(l):123-36.
(2.) Nock MK, Deming CA, Fullerton CS, Oilman SE, Goldenberg M,
Kessler RC, et al. Suicide among soldiers: A review of psychosocial risk
and protective factors. Psychiatry 2013; 76(2):97-125.
(3.) Kang HK, Bullman TA. Is there an epidemic of suicides among
current and former U.S. military personnel? Ann Epidemiol 2009;
19(10):757-60.
(4.) Statistics Canada. Canadian Forces Cancer and Mortality Study:
Causes of Death. Ottawa, ON: Statistics Canada, Catalogue no. 82-584-X,
20ll.
(5.) Belik S, Stein MB, Asmundson GJG, Sareen J. Are Canadian
soldiers more likely to have suicidal ideation and suicide attempts than
Canadian civilians? Am J Epidemiol 2010; 172(ll):1250-58.
(6.) Fear NT, Ward VR, Harrison K, Davison L, Williamson S,
Blatchley NF. Suicide among male regular UK Armed Forces personnel,
1984-2007. Occup Environ Med 2009; 66:438-41.
(7.) Black SA, Gallaway MS, Bell MR, Ritchie EC. Prevalence and
risk factors associated with suicides of army soldiers 2001-2009. Mil
Psychol 2011;23:433-5l.
(8.) Kapur N, While D, Blatchley N, Bray I, Harrison K. Suicide
after leaving the UK armed forces-a cohort study. PLoS Med 2009 Mar 3;
6(3):e26.
(9.) Griffith J. Suicide and war: The mediating effects of negative
mood, posttraumatic stress disorder symptoms, and social support among
army National Guard soldiers. Suicide Life Threat Behav 2012;
42(4):453-69.
(10.) Kaplan MS, McFarland BH, Huguet N, Newsom JT. Physical
illness, functional limitations, and suicide risk: A population-based
study. Am J Orthopsych 2007;77(l):56-60.
(11.) Zivin K, Kim HM, McCarthy JF, Austin KL, Hoggatt KJ, Walters
H, Valenstein M. Suicide mortality among individuals receiving treatment
for depression in the Veterans Affairs health system: Associations with
patient and treatment setting characteristics. Am J Public Health 2007;
97(12):2193-98.
(12.) Kaplan MS, McFarland BH, Huguet N, Valenstein M. Suicide risk
and precipitating circumstances among young, middle-aged and older male
veterans. Am J Public Health 2012; S1:S131-37.
(13.) Nock MK, Borges G, Bromet EJ, Alonso J, Angermeyer M,
Beautrais A. Crossnational prevalence and risk factors for suicidal
ideation, plans and attempts. Br J Psychiatry 2008;192(2):98-105.
(14.) Mansfield AJ, Bender RH, Hourani LL, Larson GE. Suicidal or
self-harming ideation in military personnel transitioning to civilian
life. Suicide Life Threat Behav 20ll; 41(4):392-405.
(15.) Ilgen MA, McCarthy JF, Ignacio RV, Bohnert AS, Valenstein M,
Blow FC, et al. Psychopathology, Iraq and Afghanistan service, and
suicide among Veterans Health Administration patients. J Consult Clin
Psychol 2012; 80(3):323-30.
(16.) Bossarte RM, Knox KL, Piegari R, Altieri J, Kemp J, Katz IR.
Prevalence and characteristics of suicide ideation and attempts among
active military and veteran participants in a national health survey. Am
J Public Health 2012; 102(Suppl l):S38-40.
(17.) Boulos D, Zamorski MA. Deployment-related mental disorders
among Canadian Forces personnel deployed in support of the mission in
Afghanistan, 2001-2008. CMAJ 2013; 185(ll):E545-52.
(18.) LeardMann CA, Powell TM, Smith TC, Bell MR, Smith B, Boyko
EJ, et al. Risk factors associated with suicide in current and former US
military personnel. JAMA 2013; 310(5):496-506.
(19.) Sareen J, Belik SL, Afifi TO, Asmundson GJ, Cox BJ, Stein MB.
Canadian military personnel's population attributable fractions of
mental disorders and mental health service use associated with combat
and peacekeeping operations. Am J Public Health 2008; 98(12):2191-98.
(20.) Mann JJ, Apter A, Bertolote J, Beautrais A, Currier D, Haas
A, et al. Suicide prevention strategies: A systematic review. JAMA 2005;
294(16):2064-74.
(21.) Thompson JM, MacLean MB, Van Til L, Sudom K, Sweet J, Poirier
A, et al. Survey on Transition to Civilian Life: Report on Regular Force
Veterans. Technical Report. Charlottetown, PE: Veterans Affairs Canada,
Charlottetown, and Ottawa, ON: Director General Military Personnel
Research and Analysis, Department of National Defence; January 4, 20ll.
(22.) Dominick KL, Golightly YM, Jackson GL. Arthritis prevalence
and symptoms among US non-veterans, veterans, and veterans receiving
Department of Veterans Affairs healthcare. J Rheumatol 2006;
33(2):348-54.
(23.) Ratcliffe GE, Enns MW, Belik SL, Sareen J. Chronic pain
conditions and suicidal ideation and suicide attempts: An epidemiologic
perspective. Clin J Pain 2008; 24(3):204-10.
(24.) Russell D, Turner RJ, Joiner TE. Physical disability and
suicidal ideation: A community-based study of risk/protective factors
for suicidal thoughts. Suicide Life Threat Behav 2009; 39(4):440-5l.
(25.) Scott KM, Hwang I, Chiu WT, Kessler RC, Sampson NA,
Angermeyer M. Chronic physical conditions and their association with
first onset of suicidal behavior in the world mental health surveys.
Psychosom Med 2010; 72(7):712-19.
(26.) Kline A, Ciccone DS, Falca-Dodson M, Black CM, Losonczy M.
Suicidal ideation among National Guard troops deployed to Iraq: The
association with postdeployment readjustment problems. J Nerv Ment Dis
20ll; 199(12):914-20.
(27.) Fanning JR, Pietrzak RH. Suicidality among older male
veterans in the United States: Results from the National Health and
Resilience in Veterans Study. J Psychiatr Res 2013; 47(11):1766-75.
(28.) Thompson JM, Sweet J, Poirier A, VanTil L. Suicide and
Attempt Findings in the Survey on Transition to Civilian Life:
Descriptive Analysis. Charlottetown, ON: Veterans Affairs Canada,
Research Directorate Technical Report; October 27, 20ll.
(29.) Bernal M, Haro JM, Bernert S, Brugha T, de Graaf R,
Bruffaerts R. Risk factors for suicidality in Europe: Results from the
ESEMED study. J Affect Disord 2007; 101(1-3):27-34.
(30.) Griffith JE, Vaitkus M. Perspectives on suicide in the Army
National Guard. Armed Forces and Society 2013, published online before
print February 22, 2013, doi: 10.1177/0095327X12471333.
(31.) Ilgen MA, Zivin K, Austin KL, Bohnert ASB, Czyz EK,
Valenstein M, et al. Severe pain predicts greater likelihood of
subsequent suicide. Suicide Life Threat Behav 2010; 40(6):597-608.
(32.) Bryan CJ, Hernandez AM, Allison S, Clemans T. Combat exposure
and suicide risk in two samples of military personnel. J Clin Psychol
2013; 69(l):64-77.
James M. Thompson, md, ccfp(em), fcfp, [1,2] Mark A. Zamorski, md,
mhsa, ccfp, [3] Jill Sweet, MSc, [1] Linda VanTil, dvm, MSc, [1]
Jitender Sareen, md, frcpc, [4] Robert H. Pietrzak, PhD, mph, [5] Wilma
H. Hopman, ma, [2,6] Mary Beth MacLean, ma, [1] David Pedlar, PhD [1]
Author Affiliations
[1.] Research Directorate, Veterans Affairs Canada, Charlottetown,
PE
[2.] Department of Public Health Sciences, Queen's University,
Kingston, ON
[3.] Directorate of Mental Health, Canadian Forces Health Services
Group, Ottawa, ON
[4.] Departments of Psychiatry, Psychology & Community Health
Sciences, University of Manitoba, Winnipeg, MB
[5.] United States Department of Veterans Affairs National Center
for Posttraumatic Stress Disorder, VA Connecticut Healthcare System,
West Haven, and Department of Psychiatry, Yale University School of
Medicine, New Haven, CT
[6.] Clinical Research Centre, Kingston General Hospital, Kingston,
ON Correspondence: Dr. Jim Thompson, Veterans Affairs Canada, PO Box
7700, Rm. 406 DJM Bldg, 161 Grafton St., Charlottetown, PE C1A 8M9,
E-mail: researchrecherche@vac-acc.gc.ca
Source of Funding: Government of Canada.
Conflict of Interest: Drs. Thompson, Zamorski, VanTil and Pedlar
and Ms. Sweet and Ms. MacLean are employees of the Government of Canada.
Table 1. Results of unadjusted and adjusted regression
models for past-year suicidal ideation
Variable No ideation With ideation
n = 2452 n = 206
No. (%) No. (%)
([dagger])
Physical Health Conditions
Gastrointestinal condition 315 (10.2) 68 (29.0)
Chronic pain or discomfort 1794 (62.9) 186 (83.2)
Musculoskeletal condition 1424 (48.1) 157 (65.5)
Diabetes 151 (5.0) 27 (10.0)
Respiratory condition 212 (7.5) 37 (14.5)
Cardiovascular condition 592 (20.4) 75 (32.3)
Obesity 730 (27.3) 91 (39.4)
Hearing problem 866 (27.2) 93 (38.1)
Mental Health Conditions
Depression or anxiety 558 (16.7) 160 (69.9)
Mood disorders 75 (2.1) 40 (18.6)
Anxiety disorder 273 (7.7) 96 (40.5)
Post-traumatic stress disorder 356 (8.8) 114 (40.4)
Number of physical health Mean 2.1 Mean 3.1
conditions (range 0-8) (SD 1.6) (SD 1.8)
Number of mental health Mean 0.4 Mean 1.7
conditions (range 0-4) (SD 0.8) (SD 1.3)
SF-12 physical health Mean 47.8 Mean 40.7
(SD 10.8) (SD 13.7)
SF-12 mental health Mean 53.2 Mean 33.6
(SD 9.1) (SD 11.2)
Age (years)
20-34 423 (26.1) 23 (18.7)
35-49 1101 (41.2) 131 (58.4)
50-67 928 (32.7) 52 (22.8)
Sex
Female 263 (10.7) 35 (16.8)
Male 2189 (89.3) 171 (83.2)
Marital status
Single never married 274 (15.0) 28 (18.7)
Separated/divorced/widowed 242 (8.5) 46 (22.6)
Married/common-law 1936 (76.4) 132 (58.8)
Education
Less than high school graduation 179 (6.8) 15 (4.9)
High school graduation 1033 (40.3) 93 (42.1)
Some post-secondary education 881 (36.0) 83 (44.1)
University graduate 359 (16.9) 15 (9.0)
Income
Below low income measure 137 (6.2) 24 (12.7)
Above low income measure 2315 (93.8) 182 (87.3)
Rank
Recruits 234 (15.1) 15 (10.8)
Junior non-commissioned 933 (35.1) 116 (53.2)
members ([paragraph])
Senior non-commissioned 821 (28.1) 59 (26.2)
members ([dagger][dagger])
Officers and cadets 464 (21.7) 16 (9.8)
Variable Total
n = 2658
No. (%)
Physical Health Conditions
Gastrointestinal condition 383 (11.3)
Chronic pain or discomfort 1980 (64.1)
Musculoskeletal condition 1581 (49.1)
Diabetes 178 (5.3)
Respiratory condition 249 (7.9)
Cardiovascular condition 667 (21.1)
Obesity 821 (28.0)
Hearing problem 959 (27.8)
Mental Health Conditions
Depression or anxiety 718 (19.8)
Mood disorders 115 (3.1)
Anxiety disorder 369 (9.6)
Post-traumatic stress disorder 470 (10.6)
Number of physical health Mean 2.2
conditions (range 0-8) (SD 1.6)
Number of mental health Mean 0.4
conditions (range 0-4) (SD 0.9)
SF-12 physical health Mean 47.4
(SD 11.1)
SF-12 mental health Mean 52.1
(SD 10.3)
Age (years)
20-34 446 (25.7)
35-49 1232 (42.2)
50-67 980 (32.1)
Sex
Female 298 (11.0)
Male 2360 (89.0)
Marital status
Single never married 302 (15.2)
Separated/divorced/widowed 288 (9.4)
Married/common-law 2068 (75.4)
Education
Less than high school graduation 194 (6.7)
High school graduation 1126 (40.4)
Some post-secondary education 964 (36.5)
University graduate 374 (16.4)
Income
Below low income measure 161 (6.6)
Above low income measure 2497 (93.4)
Rank
Recruits 249 (14.8)
Junior non-commissioned 1049 (36.2)
members ([paragraph])
Senior non-commissioned 880 (28.0)
members ([dagger][dagger])
Officers and cadets 480 (21.0)
Variable UOR
individual regressions
Physical Health Conditions
Gastrointestinal condition 3.59 (2.49-5.18) ***
Chronic pain or discomfort 2.91 (1.76-4.83) ***
Musculoskeletal condition 2.05 (1.41-2.97) ***
Diabetes 2.10 (1.29-3.43) **
Respiratory condition 2.10 (1.35-3.26) ***
Cardiovascular condition 1.86 (1.32-2.64) ***
Obesity 1.73 (1.24-2.42) ***
Hearing problem 1.65 (1.18-2.30) **
Mental Health Conditions
Depression or anxiety 11.59 (7.79-17.25) ***
Mood disorders 10.59 (6.51-17.22) ***
Anxiety disorder 8.22 (5.79-11.66) ***
Post-traumatic stress disorder 7.03 (5.03-9.81) ***
Number of physical health 1.46 (1.31-1.63) ***
conditions (range 0-8)
Number of mental health 2.65 (2.34-3.01) ***
conditions (range 0-4)
SF-12 physical health 0.95 (0.94-0.97) ***
SF-12 mental health 0.87 (0.86-0.89) ***
Age (years)
20-34 1.03 (0.58-1.82)
35-49 2.03 (1.40-2.95) ***
50-67 1.00
Sex
Female 1.69 (1.08-2.64) *
Male 1.00
Marital status
Single never married 1.6 (1.0-2.6)
Separated/divorced/widowed 3.44 (2.25-5.26) ***
Married/common-law 1.00
Education
Less than high school graduation 1.34 (0.60-3.00)
High school graduation 2.00 (1.04-3.72) *
Some post-secondary education 2.31 (1.21-4.40) **
University graduate 1.00
Income
Below low income measure 2.19 (1.27-3.77) ***
Above low income measure 1.00
Rank
Recruits 1.60 (0.69-3.71)
Junior non-commissioned 3.37 (1.80-6.29)*
members ([paragraph])
Senior non-commissioned 2.07 (1.08-3.96)*
members ([dagger][dagger])
Officers and cadets 1.00
Variable AORJ (95% CI)
regression with
groups of conditions,
adjusted for
socio-demographics
Physical Health Conditions
Gastrointestinal condition 1.66 (1.03-2.65) *
Chronic pain or discomfort 1.42 (0.77-2.61)
Musculoskeletal condition 0.80 (0.51-1.24)
Diabetes 1.57 (0.87-2.84)
Respiratory condition 1.06 (0.65-1.71)
Cardiovascular condition 1.46 (0.96-2.22)
Obesity 1.29 (0.89-1.86)
Hearing problem 1.03 (0.70-1.51)
Mental Health Conditions
Depression or anxiety 5.06 (2.97-8.62) ***
Mood disorders 2.91 (1.67-5.07) ***
Anxiety disorder 1.54 (0.99-2.41)
Post-traumatic stress disorder 1.31 (0.82-2.07)
Number of physical health --
conditions (range 0-8)
Number of mental health --
conditions (range 0-4)
SF-12 physical health --
SF-12 mental health --
Age (years)
20-34 1.39 (0.58-3.36)
35-49 1.59 (0.97-2.60)
50-67 1.00
Sex
Female 1.14 (0.65-2.00)
Male 1.00
Marital status
Single never married 1.92 (1.05-3.49) *
Separated/divorced/widowed 2.50 (1.53-4.07) ***
Married/common-law 1.00
Education
Less than high school graduation 0.59 (0.23-1.54)
High school graduation 1.14 (0.57-2.30)
Some post-secondary education 1.15 (0.57-2.32)
University graduate 1.00
Income
Below low income measure 1.58 (0.84-2.98)
Above low income measure 1.00
Rank
Recruits 1.11 (0.37-3.35)
Junior non-commissioned 0.96 (0.30-3.13)
members ([paragraph])
Senior non-commissioned 0.72 (0.26-2.03)
members ([dagger][dagger])
Officers and cadets 1.00
Variable AOR ([section]) (95% CI)
regression with
numbers of
conditions,
adjusted for
socio-demographics
Physical Health Conditions
Gastrointestinal condition --
Chronic pain or discomfort --
Musculoskeletal condition --
Diabetes --
Respiratory condition --
Cardiovascular condition --
Obesity --
Hearing problem --
Mental Health Conditions
Depression or anxiety --
Mood disorders --
Anxiety disorder --
Post-traumatic stress disorder --
Number of physical health 1.22 (1.05-1.42) **
conditions (range 0-8)
Number of mental health 2.32 (2.01-2.68) **
conditions (range 0-4)
SF-12 physical health --
SF-12 mental health --
Age (years)
20-34 1.30 (0.54-3.13)
35-49 1.49 (0.94-2.37)
50-67 1.00
Sex
Female 1.35 (0.79-2.33)
Male 1.00
Marital status
Single never married 1.97 (1.10-3.52) *
Separated/divorced/widowed 2.51 (1.51-4.17) **
Married/common-law 1.00
Education
Less than high school graduation 0.59 (0.24-1.48)
High school graduation 1.00 (0.50-1.99)
Some post-secondary education 1.08 (0.54-2.16)
University graduate 1.00
Income
Below low income measure 1.58 (0.84-3.00)
Above low income measure 1.00
Rank
Recruits 1.42 (0.48-4.27)
Junior non-commissioned 1.40 (0.66-2.99)
members ([paragraph])
Senior non-commissioned 1.33 (0.64-2.78)
members ([dagger][dagger])
Officers and cadets 1.00
Variable AOR ([parallel])
(95% CI) regression
with SF-12 component
summaries,
adjusted for
socio-demographics
Physical Health Conditions
Gastrointestinal condition --
Chronic pain or discomfort --
Musculoskeletal condition --
Diabetes --
Respiratory condition --
Cardiovascular condition --
Obesity --
Hearing problem --
Mental Health Conditions
Depression or anxiety --
Mood disorders --
Anxiety disorder --
Post-traumatic stress disorder --
Number of physical health --
conditions (range 0-8)
Number of mental health --
conditions (range 0-4)
SF-12 physical health 0.98 (0.96-0.99) *
SF-12 mental health 0.88 (0.87-0.89) ***
Age (years)
20-34 0.84 (0.34-2.08)
35-49 0.98 (0.61-1.61)
50-67 1.00
Sex
Female 1.01 (0.56-1.84)
Male 1.00
Marital status
Single never married 2.22 (1.14-4.30) *
Separated/divorced/widowed 1.60 (0.96-2.68)
Married/common-law 1.00
Education
Less than high school graduation 0.50 (0.19-1.33)
High school graduation 0.92 (0.44-1.93)
Some post-secondary education 1.06 (0.50-2.25)
University graduate 1.00
Income
Below low income measure 1.06 (0.56-2.01)
Above low income measure 1.00
Rank
Recruits 1.61 (0.55-4.75)
Junior non-commissioned 1.50 (0.74-3.04)
members ([paragraph])
Senior non-commissioned 1.37 (0.65-2.91)
members ([dagger][dagger])
Officers and cadets 1.00
([dagger]) All numbers represent unweighted sample sizes, and
all percentages represent weighted column percent
(percentage of those with or without ideation who had the
condition or characteristic); for number of health
conditions and health conditions and SF-12 component
summaries, mean and standard deviation shown.
([double dagger]) All variables shown were entered into the
model at once to adjust for health conditions and the socio-
demographic variables, df=2627, F(24,2604)=13.64,
probability of greater F < 0.001; excludes numbers of health
conditions and the SF-12 component summaries.
([section]) All variables shown were entered into the model at
once to adjust for numbers of conditions and the socio-
demographic variables, df=2655, F(14,2642)=22.8, probability
of greater F < 0.001; excludes individual health conditions
and the SF-12 component summaries.
([parallel]) All variables shown were entered into the model
at once to adjust for the SF-12 component summaries and the
socio-demographic variables, df=2658, F(14,2642) = 26.0,
probability of greater F < 0.001; lower scores on the
Physical Component Summary and the Mental Component Summary
reflect worse health; excludes individual health conditions
and numbers of conditions.
* p<0.05, ** p<0.01, *** p<0.001
([paragraph]) Private, Corporal, Master Corporal; Ordinary, Able,
Leading and Master Seaman.
([dagger][dagger]) Sergeant, Warrant Officer, Master Warrant
Officer, Chief Warrant Officer, Petty Officer and Chief
Petty Officer.
UOR = Unadjusted odds ratio for past-year suicidal ideation,
reference group without the condition; CI, confidence
interval; AOR, odds ratio adjusted for covariates.
"-", variable not included in the model.
