A spatial epidemiological analysis of self-rated mental health in the slums of Dhaka

Table 3 Global bivariate Moran's I values for the three nearest neighbours

Scale level	Health-Determining Factor	Beguntila	Bishil/Sarag
	WHO-5 scores ~	Young adults n = 115	Young adults n = 170	Females n = 104	Males n = 122
Global univariate Moran's I for WHO-5 scores		0.16*	0.12*	0.12*	0.19**
Neighbourhood level physical environment	'Natural Environment' Flood non-affectedness	-0.19*** .	-0.16* 0.13	. 0.13*	-0.21** .
Household level physical environment	Housing quality Basic services Household wealth Sanitation Housing sufficiency Housing durability	0.13* . . . . .	0.19*** . . . . .	0.14** . . . . .	0.3* . . 0.18* . .
Household level social environment	Population density Job satisfaction Income generation	. . .	. . 0.1*	. . .	. . .
Individual level	Smoking behaviour Environmental HK Personal HK Community member Using bed net Education Married Migrant Age Gender	. . . . . . . . --- 0.12*	. 0.09* . . . . . . --- .	. . . . . 0.11* . . -0.12* ---	. 0.13* . . . . . . . --- .

Significance levels: < 0.001 '***', < 0.01 '**', < 0.05 '*', > 0.05 '.', not applicable '---'
HK: Health knowledge
The table displays health-determining factors that are significantly (p < 0.05) spatially correlated with mental health (WHO-5 scores) of those population groups and slums in which strongest global spatial clustering of WHO-5 scores were found (cf. Table 2). Note that the WHO-5 scores among males in Bishil/Sarag are clustered most strongly, and there is a strong spatial correlation with 'natural environment' and housing quality in this population group.

ISSN: 1476-072X