Modeling predictors of risky drug use behavior among male street laborers in urban Vietnam

BMC Public Health

Table 1 The constructs of the original and modified IMB model

Constructs	References	Number of items	Scale	Crobach’s α
The original model
*HIV preventive information*	[21, 22]	7	True/false	0.63
Transmission information		3	True/false	0.66
Heuristic information		4	True/false	0.61
*HIV preventive motivation*	[23–25]	21	5-point semantic¶	0.91
Attitudes		7	5-point semantic¶	0.75
Social norms		7	5-point semantic§	0.83
Intentions		7	5-point semantic†	0.81
*HIV preventive behavioral skills*	[23, 26]	5	5-point semantic#	0.86
Skill 1 (preparation)		3	5-point semantic#	0.76
Skill 2 (practice)		2	5-point semantic#	0.91
The modified model
*3 constructs of the original model*
*Psychosocial stress‡*		4	Different scales	0.71
Mobility index‡	[27]	2	-	-
Social isolationŸ	[28]	6	5-point	0.74
Alcohol use¥	[29]	2	-	0.60
Depressionƒ	[30–34]	10	4-point	0.88

Note: ¶ scale from 1 (negative evaluation) to 5 (positive evaluation); §scale from 1 (negative evaluation) to 5 (positive evaluation); †scale from 1 (very unlikely) to 5 (very likely); #scale from 1 (very hard) to 5 (very easy); ‡ the ratio of the number of migratory cities to years of total migration. (−) Not applicable as it is a ratio; Ÿscale from 0 (not at all) to 4 (almost always) during the past 4 weeks; ¥ a composite of the number of standard drinks and frequency of use over the past 4 weeks; ƒscale from 0 (rarely or none of the time) to 3 (most or all of the time) during the past week.
‡because the four indicators were associated with stresses given our qualitative findings [20, 35] and statistic parameters (Pearson’s correlation coefficients from the current quantitative data), they were formed to serve as a latent construct of psychosocial stress (α = .71).
All of the above measures have been adapted from the tools by other authors as well as from our qualitative research.

ISSN: 1471-2458