Table 4.

Effect of POW status on overall and heart disease mortality, men younger than 65 in 1900

	Entire Sample			Known Cause of Death
	Overall Mortality			All Causes	Heart Disease			Valvular	Stroke	Ischemic
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
Dummy = 1 if
Non-POW
POW, captured early or late	0.991 (0.040)	1.016 (0.041)	1.017 (0.042)	1.031 (0.058)	1.041 (0.088)	1.042 (0.090)	1.020 (0.088)	4.970** (1.870)	1.058 (0.161)	1.244 (0.238)
POW, no exchange	1.089* (0.043)	1.110** (0.044)	1.110* (0.047)	1.137* (0.062)	1.211* (0.098)	1.191* (0.099)	1.127 (0.099)	9.028** (2.995)	1.296† (0.182)	1.845** (0.310)
Test of Equality of
POW, captured early or late
POW, no exchange
χ2(1)	3.16	2.73	2.62	1.76	1.86	1.44	0.79	2.37	1.09	2.77
Prob > χ2	0.076	0.100	0.105	0.184	0.173	0.230	0.374	0.124	0.297	0.096
Pre- and Enlistment SES Controls	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Postwar SES Controls	N	Y	Y	N	N	Y	Y	N	N	N
Wartime Health and Scurvy Controls	N	N	Y	N	N	N	Y	N	N	N
Number of Deaths	9,394	9,394	9,394	4,879	2,124	2,124	2,124	54	649	376
Observations	9,394	9,394	9,394	4,879	4,879	4,879	4,879	4,879	4,879	4,879

Notes: Regressions are Eq. 1 in the text and are from a Cox regression in which the baseline hazard is calendar years lived after 1900. All regressions control for age in 1900. The sample consists of all veterans alive and on the pension rolls in 1900, with full information on dates of captivity and age in 1900. The table gives hazard ratios. Standard errors are in parentheses.

^†p ≤ .10;

^*p ≤ .05;

^**p ≤ .01