Table 1. Impact of misclassification between schizophrenia (disorder A) and bipolar disorder (disorder B) on estimation of genetic parameters from recurrence risks in first-degree relatives.

				rgT=0			rgT=0.25			rgT=0.5
MTA	MTB	KDA	KD	h2DA	h2DB	rgD	h2DA	h2DB	rgD	h2DA	h2DB	rgD
0	0	1.00	1.00	76	76	0	76	76	25	76	76	50
5	5	1.00	1.00	72	72	21	73	73	39	74	74	59
10	10	1.00	1.00	68	68	37	69	69	51	71	71	67
15	15	1.00	1.00	65	65	50	66	66	61	69	69	74
20	20	1.00	1.00	62	62	62	64	64	70	67	67	80
30	30	1.00	1.00	56	56	82	59	59	86	63	63	91
40	40	1.00	1.00	52	52	95	56	56	96	61	61	98
50	50	1.00	1.00	51	51	100	55	55	100	60	60	100
0	5	1.05	0.95	74	75	11	74	75	32	75	75	54
0	10	1.10	0.90	71	74	20	72	74	38	74	74	58
0	15	1.15	0.85	69	73	28	71	73	44	73	73	62
0	20	1.20	0.80	67	71	34	69	71	48	72	71	65
0	30	1.30	0.70	65	69	45	67	69	57	71	69	70

Parameters follow those used in Table 3 of Kendler.⁵ All values are expressed as percentages. The true disease prevalences are assumed to be 1% for both schizophrenia and bipolar disorder, K_TA=K_TB=1%. True recurrence risks to first-degree relatives are λ_TA=λ_TB=8.0. These parameters equate to true heritabilities on the liability scale of h_TA²=h_TB²=0.76. M_TA is the proportion of true schizophrenia cases misclassified as bipolar disorder and M_TB is the proportion of true bipolar disorder cases misclassified as schizophrenia. The true genetic correlation between the disorders is r_gT=0, 0.25,0.5. The estimated parameters based on diagnosed prevalences and recurrences risks have subscript D.