The economic impact of Marfan syndrome: a non-experimental, retrospective, population-based matched cohort study

Table 4 Average treatment effects for the treated (excess costs) in € (per capita, in 2008)

	Sickness fund perspective					Societal perspective
	Marfan	Control	ATT^a		AI SE^b	Marfan	Control	ATT^a		AI SE^b
Direct medical costs	4024	1695	2330	*******	273	4105	1739	2366	*******	278
Outpatient treatment	780	512	268	***	35	800	531	269	***	35
Pharmaceuticals	385	241	145	**	46	349	222	127	**	39
Care by non-physicians	1315	472	843	***	163	1353	501	851	***	165
Devices and medical appliances	142	76	66	***	18	122	65	57	***	15
Inpatient treatment	1337	379	958	***	164	1413	403	1010	***	173
Rehabilitation	66	14	52	***	15	67	14	53	***	16
Medical services	<1	1	−1		5	<1	1	−1		2
Direct non-medical costs	392	226	167	*******	36	7431	1556	5875	*******	116
Administration	192	192	0		0	192	192	0		0
Sick leave compensation	161	18	143	***	34	n/a	n/a	n/a		n/a
Travel expenses	39	15	24	***	7	39	15	24	**	9
Other non-medical services	<1	<1	<1		<1	<1	<1	<1		<1
Informal family care	n/a	n/a	n/a		n/a	7200	1349	5851	***	116
Indirect costs	n/a	n/a	n/a		n/a	10,329	2842	7487	*******	718
Total costs	4416	1920	2496	*******	286	21,865	6137	15,728	*******	824.4
Indicators for healthcare utilisation
Sick leave days	9.23	5.02	4.21	***	1.11
Physician contacts	10.11	7.28	2.83	***	0.26
Average length of stay	3.06	1.21	1.85	***	0.52
Inpatient stays	0.39	0.18	0.21	***	0.04
Prescriptions	8.70	6.52	2.18	***	0.49
Average prescription cost	27.58	23.04	4.55		3.61
MRT/CT imaging	0.37	0.11	0.26	***	0.05

^aaverage treatment effect for the treated represents excess resource utilisation attributable to Marfan.
^bAbadie–Imbens standard errors take the uncertainty of the matching process into consideration [57].
**< 0.01 ***< 0.001.

ISSN: 1750-1172