Table.

Deaths on day 4 of each month as a percentage of deaths on days 3, 4, and 5 (with number of deaths on days 3, 4, and 5 in parentheses)

	Day 4 deaths			Two sided P values
	White Americans	Chinese or Japanese Americans		Binomial*	Fisher's exact test†
1989-98
Total coronary deaths:	33.3 (60 839)	33.9 (1391)		0.652	0.664
Inpatients	33.8 (20 658)	36.5 (540)		0.133	0.211
Inpatients, “chronic”‡	33.8 (15 084)	38.9 (404)		0.022	0.040
1999-2001
Total coronary deaths:	33.2 (18 598)	30.9 (505)		0.264	0.293
Inpatients	32.6 (6 061)	31.8 (195)		0.710	0.873
Inpatients, “chronic”‡	32.2 (4 273)	34.6 (133)		0.822	0.626
1969-1988
Total coronary deaths:	33.4 (120 301)	34.9 (992)		0.317	0.353
“Chronic”‡	33.5 (98 257)	35.0 (734)		0.353	0.399
1969-2001
Suicides	33.5 (11 150)	27.8 (198)		0.110	0.100
Accidents	33.7 (30 508)	36.2 (390)		0.259	0.326

^*Binomial test of null hypothesis that, for each Chinese or Japanese death on days 3, 4, or 5 of a month, there is a 33% probability that it occurs on the day 4. 

^†Fisher's exact test with data separated by race and by whether death occurred on day 4 or on adjacent days. 

^‡The heart diseases selected by Phillips et al¹—ICD-8 410-414; ICD-9 410-414; ICD-10 I20-I24, I25.1-I25.9.