Post-hospital medical respite care for homeless people in Denmark: a randomized controlled trial and cost-utility analysis

Table 3 Cost computation regression models comparing control and intervention groups (average per person)

	Model 1 3 months €	Model 1 6 months €	Model 1 12 months €	Model 2 3 months €	Model 2 6 months €	Model 2 12 months €
Constant	1721	−603	− 2865	131	− 3086	− 4806
Intervention group	−4761	−8515*	−12,603*	−4328	−8161	−10,687
Costs before	0.15	0.25*	0.38*	0.08	0.17	0.25*
Education	–	–	–	1318	3266	4071
Charlson Comorbidity Index	–	–	–	1526*	2177*	5215*
Functionally homeless	–	–	–	9139*	7826	4912
Homeless other	–	–	–	256	− 1297	5464
Homeless missing	–	–	–	− 2787	− 4822	− 4867
R²	0.0575	0.1156	0.2038	0.1698	0.2055	0.3259

*p < 0.05
Model 1 is a difference in difference analysis and model 2 is difference in difference adjusted for level of education, Charlson Comorbidity Index and type of homelessness

ISSN: 1472-6963