A genetic and epidemiological study of cleft lip and cleft palate in Hawaii

Abstract

The purpose of this study was (1) to obtain unbiased estimates of cleft incidences in the major racial groups in Hawaii, (2) to evaluate the extent and nature of racial differences in incidence, (3) to determine how the segregation of cleft defects varies among racial groups, and (4) to estimate the heritability of cleft defects using a multifactorial model of quasicontinuity. Cleft cases were ascertained through birth and death certificates and records of the Crippled Children Branch (Hawaii Health Department) and major Honolulu hospitals. The reference population which provided the basis for the analysis of incidence was composed of live births in Hawaii from 1948 to 1966, as compiled by Drs. Morton, Chung, and Mi. Live birth certificates provided demographic data on all individuals in the reference population. Variables included race and age of each parent, sex of child, birth order, year of birth, and occupation and military status of father. Families of probands were interviewed for additional data on types and numbers of affected and normal relatives and parental consanguinity. The ascertainment of CL(P) was estimated to be 97% complete, and that of CP to be 93% complete. The danger of relying exclusively on birth certificates for the ascertainment of cases was made clear by our findings that only 80.7% of CLP, 66.7% of CL, and 47.7% of CP probands born in 1948-1966 had their defects noted on their birth certificates. A number of factors were found to be related to the ascertainment probabilities of CP and CL(P). In the case of CP, ascertainment appeared more likely when the individual had three or more additional malformations and less likely when his father was actively engaged in military service. The case of CL(P) was more complicated. The probability of ascertainment of CL(P) apparently increased when the defect extended to the palate and when any additional malformation was present; it seemed to decrease with year of birth, age of father, death in infancy, and father in the military. Significantly, however, no racial biases in ascertainment were detected after adjustment for the effects of non-racial factors. After correcting for ascertainment biases, the estimated overall incidence of CP in Hawaii was .78 per thousand live births. In the pure Caucasian group, the estimated incidence was .50; in the pure Japanese group, it was .74. Incidences in excess of one per thousand were found in groups with Hawaiian ancestry. Regression analysis showed a definite association between Hawaiian parentage and high incidence of CP. This strong Hawaiian effect could not be attributed to maternal racial factors or to interracial hybridization. Orientals as a whole did not appear to have a significantly higher incidence of CP than Caucasians. For CL(P), the corrected incidence was estimated at 1.05 per thousand in Caucasians, 1.92 in Japanese, and 1.28 in the general population. Unlike in the case of CP, Hawaiian ancestry was not associated with high incidence of CL(P). Instead, the Oriental groups, particularly the Japanese and Filipinos (1.56), presented the highest risks. No maternal or hybridity racial effects were detected for any of the Oriental groups. The effects of non-racial factors (father's age, mother's age, birth order, year of birth, father's occupational and military status) on incidence were tested by separate regression analyses in the Japanese and in the Caucasian racial . groups. No consistent relationship between incidence and any non-racial variable was detected for either cleft type. However, CP incidence did appear to increase with lower occupational status in the Caucasian group, although not in the Japanese group. Estimates of mean segregation frequencies for CP were 2.2% in families of Hawaiian ancestry and 1.3% in all other families. Although the difference was not statistically significant, the higher risk in Hawaiian families would be expected under the multifactorial hypothesis which equates risk approximately to the square root of the population incidence. Heritability of CP was estimated to be 69%, which possibly indicates that additive genetic factors are very important in the etiology of CP. Mean segregation frequencies for CL(P) were 6.1% in pure Japanese, 6.4% in pure Filipinos, and 4.2% in all other groups. The higher risks in Japanese and Filipinos may reflect their higher incidences of CL(P). Overall heritability of CL(P) was estimated at 86%, with no significant differences among Japanese, Filipinos, and all other groups combined. The weight of evidence now seems to favor a multifactorial hypothesis of CL(P) inheritance, and the present estimates of heritability would suggest a large additive hereditary component.