Table 4.

Model Fitting for Associations of Hurricane-Related Stressors With Estimated 30-Day DSM-IV Anxiety-Mood Disorders^a

	New Orleans Metro (n=594)									Remainder of Sample (n=449)
	PTSD			Other MMI or SMI			Any MMI or SMI			PTSD			Other MMI or SMI			Any MMI or SMI
Modelb	χ2	P Value	df	χ2	P Value	df	χ2	P Value	df	χ2	P Value	df	χ2	P Value	df	χ2	P Value	df
M1	59.5	<.001	17	21.8	.19	17	44.4	<.001	17	74.5	<.001	18	42.6	<.001	18	83.0	<.001	18
M2-M1	31.2	<.001	7	45.8	<.001	7	59.0	<.001	7	36.6	<.001	7	13.7	.06	7	22.2	<.001	7
M3-M1	35.6	<.001	3	35.0	<.001	3	57.8	<.001	3	10.8	<.001	2	7.8	.02	2	16.4	<.001	2
M4-M3	47.7	.37	45	60.6	.06	45	34.8	.86	45	22.4	>.99	45	38.8	.73	45	31.8	.93	45

Abbreviations: M, model; Metro, metropolitan area; MMI, mild/moderate mental illness; PTSD, posttraumatic stress disorder; SMI, serious mental illness.

^aEstimates of anxiety-mood disorders were based on the K6 and Trauma Screening Questionnaire scales. See the “Methods” section for details.

^bThe predictors in M1 consist of the sociodemographic variables in Table 2. The predictors in M2 include those in M1 plus 7 dummy variables for the number of stressors to which the respondent was exposed. The predictors in M3 include those in M1 plus a single dummy variable for exposure to 1 or more stressors and either dummy variables for physical illness/injury and physical adversity (in the New Orleans metro) or a dummy variable for property loss (in the remainder of the sample). The predictors in M4 include those in M3 plus a separate dummy variable for each of the 45 logically possible 2-way interactions among the 10 stressors.