Suicide attempts among activated soldiers in the U.S. Army reserve components

BMC Psychiatry

Table 2 Multivariable associations of socio-demographic characteristics with suicide attempt among activated officers in the U.S. Army reserve components^a

	OR	(95% CI)	Cases (N)	Total (N)^b	Rate^c	Pop %^d	SRE^e
Gender
Male	1.0	–	30	1,858,430	19	84.7	19
Female	3.3*	(1.7–6.1)	17	336,017	61	15.3	61
χ²₁	13.7*
Age at Army Entry (years)
< 21	0.9	(0.5–1.8)	22	1,158,822	23	52.8	24
21–24	1.0	–	15	694,415	26	31.6	25
≥ 25	1.3	(0.6–3.0)	10	341,210	35	15.5	34
χ²₂	0.8
Current Age (years)
≤ 29	0.7	(0.3–2.0)	7	231,407	36	10.5	38
30–34	1.0	–	10	280,610	43	12.8	46
35–39	0.4	(0.2–1.1)	8	486,008	20	22.1	21
≥ 40	0.4*	(0.2–0.9)	22	1,196,422	22	54.5	21
χ²₃	5.7
Race/Ethnicity
White	1.0	–	38	1,688,438	27	76.9	29
Non-White	0.6	(0.3–1.3)	9	506,009	21	23.1	18
χ²₁	1.5
Education
≤ College	1.0	–	31	1,511,631	25	68.9	24
> College	1.2	(0.7–2.4)	16	682,816	28	31.1	30
χ²₁	0.5
Marital Status
Never/Previously Married	1.0	–	19	676,219	34	30.8	28
Currently Married	0.9	(0.5–1.6)	28	1,518,228	22	69.2	24
χ²₁	0.2
Total			47	2,194,447	26	100.0	–

^aThe sample of activated Reserve Component officers (n = 47 cases, 10,972 control person-months) is a subset of the total Reserve Component sample (n = 70,970 person-months) from the Army STARRS Historical Administrative Data Study (HADS). All control person-months were assigned a weight of 200 to adjust for under-sampling. The analysis included a dummy predictor variable for calendar month and year to control for secular trends
^bTotal includes both cases (i.e., suicide attempters) and weighted control person-months
^cRate per 100,000 person-years, calculated based on n₁/n₂, where n₁ is the unique number of soldiers within each category and n₂ is the annual number of person-years, not person-months, in the population
^dPop % Percentage of the activated Reserve Component officer population
^eSRE Standardized risk estimate (suicide attempters per 100,000 person-years) was calculated assuming other predictors were at their sample-wide means
*p < 0.05

ISSN: 1471-244X