Table 2.

Percentage breakdown of the demographics and the drugs used for the patients in the datasets, all of which were tested as covariates

	M	V		M	V		M	V
	%	%	Diagnosis	%	%	HSCT	%	%
Age at HSCT, years
0→1	16	19	Immunodeficiencies	43	40	1st	85	88
1→2	21	16	SCID	26	24	2nd	13	11
2→5	23	21	Wiskott‐Aldrich	4	7	3rd	1	1
5→10	24	31	CGD	4	8	GvHD
10→	16	13	Leukemia	30	23	Reported	32	60
Sex			ALL	14	11	I	12	33
Male	37	32	AML	11	11	II	12	20
Female	63	68	HLH	11	7	III	6	5
Stem cells			Anemia	7	0	IV	2	1
Bone marrow	47	36	Autoimmune	3	0	Conditioning
Peripheral blood	38	37	Lymphomas	2	0	Fludarabine	21	73
Cord blood	15	27	Viruses			Cyclophosphamide	44	16
Combinations	1	0	Cytomegalovirus			Melphalan	30	23
Donor type			Positive	32	16	Busulphan	24	41
Matched	63	52	Negative	67	81	Treosulphan	21	24
Sibling	27	19	Unknown	1	3	Alemtuzumab	50	40
Family	5	7	Epstein‐Barr virus			ATG	3	16
Unrelated	31	27	Positive	26	16	Anti‐CD45	4	3
Mismatched	32	37	Negative	38	64	Total body irradiation	14	8
Sibling	1	0	Unknown	37	3	None	13	5
Family	2	1	Adenovirus			Prophylaxis
Unrelated	29	36	Positive	33	–	Cyclosporine	88	88
Haploidentical	4	3	Negative	67	–	Methotrexate	21	16
Autologous	1	8				Mycophenolate	50	68

ALL, acute lymphoblastic leukemia; AML, acute myeloblastic leukemia; ATG, antithymocyte globulin; CGD, chronic granulomatous disease; GvHD, graft‐vs.‐host disease; HLH, hemophagocytic lymphohistiocytosis; SCT, stem cell transplantation; M, model‐building dataset (n = 319), used for model building and covariate analysis; SCID, severe combined immunodeficiency syndrome; V, validation dataset (n = 75), used for assessing the predictive ability of the model.

Positive for cytomegalovirus, Epstein‐Barr virus, or adenovirus was defined as detectable virus post‐transplant.