Familial Mental Retardation

Size: px

Start display at page:

Download "Familial Mental Retardation"

Melanie Warren
6 years ago
Views:

1 Behavior Genetics, Vol. 14, No. 3, 1984 Familial Mental Retardation Paul L. Nichols ~ Received 18 Aug Final 2 Feb Familial patterns of mental retardation were examined among white and black children in the NINCDS Collaborative Perinatal Project population. Among whites, the mildly retarded children had more affected relatives than did the severely retarded, consistent with the traditional two-group theory of mental retardation. In blacks, where differences in family patterns between the mildly and the severely retarded were less clear, most retardation appeared to be of the cultural-familial type. KEY WORDS: mental retardation; severe retardation; mild retardation; family studies; Collaborative Perinatal Project. INTRODUCTION A two-group theory is often used to describe mental retardation in the population (Zigler, 1967). Severely retarded children, with IQs below 50, are characterized by central nervous system damage, an average socioeconomic status, and few affected relatives. Mildly retarded children (IQ, 50-69) generally have no pathological signs, a low socioeconomic status, and many affected relatives. Differences between relatives of the two types of retardates is said to be one of the best lines of evidence in support of the two-group theory (Jensen, 1969). Reports of relatives of the retarded are generally old (Akesson, 1961; Penrose, 1939; Roberts, 1940, 1952). Roberts' (1952) graphical presentation of the IQs of siblings of retardates with IQs above and below 50 has been widely reproduced (Jensen, 1969; Vandenberg, 1968). Kamin (1974), however, pointed out a methodological problem in that study, noting that Roberts actually classified his subjects based on their siblings' 1National Institute of Neurological and Communicative Disorders and Stroke, Bethesda, Maryland, /84/ /09 Plenum Publishing Corporation

162 Nichols rather than their own IQs. No subsequent reports of IQ distributions among siblings of retardates have been published. Nor have any reports been published for black populations.

2 162 Nichols rather than their own IQs. No subsequent reports of IQ distributions among siblings of retardates have been published. Nor have any reports been published for black populations. In this study, the IQ distribution and frequency of retardation are examined for siblings of severely and mildly retarded children in white and black samples. In addition, data are presented for half-siblings, cousins, and twins of the two retarded groups. METHODS The data reported are part of a study of mental retardation from the Collaborative Perinatal Project of the National Institute of Neurological and Communicative Disorders and Stroke? This project (or NCPP), described in detail by Broman et al. (1975) is a large multidisciplinary study relating prenatal and perinatal events to adverse outcomes of pregnancy. At age 7 years, the children in the project were given a battery of psychological tests, including the Wechsler Intelligence Scale for Children (WISC). Four verbal and three performance subtests were used. Children whose IQs were too low to be measured with the WISC were given other tests appropriate to their ability such as the Stanford-Binet Intelligence Scale, Vineland Social Maturity Scale, Cattell Infant Intelligence Scale, and Bayley Scale of Infant Mental Development. Children with IQ scores below 50 were classified as severely retarded, and those with IQs between 50 and 69 were classified as mildly retarded. To include as many retarded children as possible in the cohort, IQ levels were estimated for children who were rated untestable or who were institutionalized at 7 years of age. Judgments were based on detailed clinical descriptions and prior examinations. Overall, 18% of the severely retarded and less than 1% of the mildly retarded were classified in this way. These percentages did not differ between the white and the black samples. Among 17,432 white children, 0.5% were severely (IQ, <50) and 1.2% mildly (IQ, 50-69) retarded. Most (72%) of the severely retarded children were found, when examined at age 7 years, to have severe central nervous system pathology, manifested by one or more of the following: Down's syndrome, other syndromes, post traumatic deficit, CNS malformations, cerebral palsy, epilepsy, and sensory deficits. In contrast, only 14% of the mildly retarded children had any of these conditions. Among 19,419 black children, 0.7% were severely retarded and 4.6% mildly retarded. Again, many more of the severely retarded than the mildly retarded showed evidence of central nervous system pathology (54 2 Broman, S. H., Nichols, P. L., Shaughnessy, P. W., and Kennedy, W. A. (1984). Mental Retardation: A Longitudinal Study (in preparation).

3 Familial Mental Retardation Siblings of Siblings of Mildly /,,q,, Severely / Retarded.30 Retarded /,, = /.10 0 f I I 1... I 1 I I Seven Year IQ Fig. 1. IQ distribution for siblings of severely retarded (IQ, <50) and mildly retarded (IQ, 50-69) white children. vs. 6%). The frequency of retardation (IQ, <70) was determined for relatives of these retardates who had also participated in the NCPP. In order to use all the data possible in comparing the frequency of retardation between relatives of severely and relatives of mildly retarded children, in families with two affected individuals each was counted as both an index and a secondary case. This procedure is unbiased in studies such as the NCPP, in which families with more than one affected child were not more likely to be identified than families with only one affected. RESULTS Siblings The severely retarded white children had 20 siblings in the study. None of these siblings was retarded, and their mean IQ of (SD, 12.1) was near that of the NCPP white population. The 58 siblings of the mildly retarded white children showed a different pattern: 12 (20.7%) were retarded, and their mean IQ was only 84.8 (SD, 18.1). Figure I shows that the distribution of these siblings' IQ scores is clearly shifted downward. The apparent bimodality of the curve for siblings of mildly retarded was unrelated to socioeconomic status, sex, or siblings' IQ and therefore most likely was a random fluctuation due to small sample size. In contrast to the whites, the siblings of severely and mildly retarded black children did not show a different pattern. Retardation was common among siblings of both the severely and the mildly retarded (20.8% of 40 and 21.1% of 327, respectively), and the mean IQs of the siblings were very similar (77.8 and 78.1, respectively). These data suggest a less clear distinction between the severely and the mildly retarded groups, based

$164 Nichols.40 Siblings of Mildly Retarded,~,.30 / \ \..20 Siblings of Severely ~ o.10 Retarded ll'~" ~///,/" ~ ~ Fig. 2.$

4 164 Nichols.40 Siblings of Mildly Retarded,~,.30 / \ \..20 Siblings of Severely ~ o.10 Retarded ll'~" ~///,/" ~ ~ Fig. 2. c,- S g'/" I I I I 0 ~" Seven Year IQ IQ distribution for siblings of severely retarded (IQ, <40) and mildly retarded (IQ, 40-59) black children. on an IQ cutoff of 50, for blacks than for whites. Part of the overlap may have been caused by too high an IQ cutoff, because IQ distributions of black populations are generally shifted downward. In support of this hypothesis, only 6 of 37 (16%) "severely" retarded black children with IQs between 40 and 49 had evidence of central nervous system pathology, compared to 63 of 91 (69%) with IQs below 40. Also the mean socioeconomic index score (Myrianthopoulos and French, 1968) of the latter group was significantly higher (37.6 vs. 23.5, P < 0.001, compared to 38.6 in the total black population). For an additional analysis, the IQ cutoffs for blacks were shifted, so that severe retardation was defined as an IQ below 40 and mild retardation as an IQ between 40 and 59. The distributions of the IQs of the siblings of the two new groups did differ slightly (Fig. 2). The 20 siblings of the severely retarded had a mean IQ of 77.8, with a large standard deviation of The 73 siblings of the mildly retarded had a similar mean (76.0) but a lower standard deviation (13.7). The greater variability among the siblings of the severely retarded children resulted from two brothers with IQs of 25 and 38; both had epilepsy, Without this family, the difference between the two group means (83.1, SD 13.4, vs. 76.0) would have been nearly significant (P = 0.055). The familial incidence of severe retardation is generally thought to be confined largely to specific abnormalities, none of which was present among the 20 study siblings of the severely retarded white children. There was, however, limited information available for nonstudy siblings as well, because the mothers were asked during both prenatal and 7-year interviews for information on retardation in all family members. The frequency of retardation was determined for all siblings of the 92 severely retarded

5 Familial Mental Retardation 165 Table I. Reported Retardation Among Siblings of Severely Retarded White Children Study child's diagnosis Phenylketonuria Aminoaciduria Cerebral palsy Epilepsy Carnosinemia Epilepsy Chromosomal translocation Undifferentiated Undifferentiated Sibling's diagnosis Phenylketonuria Aminoaciduria (died age 9 months) Cerebral palsy, epilepsy, severe retardation Epilepsy, severe retardation, institutionalized Epilepsy, severe encephalopathy Hydrocephaly Postmeningitic encephalopathy Educably retarded, in special classes In special classes, could print name at age 10 years white children. Excluded were any siblings 1 year of age or less, except one child who died at age 9 months, apparently with the same aminoaciduria as his brother in the cohort. Based on information in these interviews, there was fairly good evidence that 9 of 196 siblings were retarded. The diagnoses of these nine sibling pairs are shown in Table I. This frequency of retardation (4.6%) was higher than the total frequency in the white study population (1.7%) but should be interpreted with caution because the diagnoses were based on mothers' reports. Most of the diagnoses, however, seemed straightforward. There were several instances of similar conditions for siblings within a family. The two cases in which the study child's retardation was undifferentiated had relatively high IQs (46 and 47) and appeared to be cultural-familial. Half-Siblings Study children of mothers who reported different fathers were considered half-siblings. The severely retarded white children had no halfsiblings in the study, and the mildly retarded white children had only seven. None of the seven was retarded, and their mean IQ was 89.1 (SD, 9.4). The severely retarded black children had only four half-siblings in the study. One was retarded (25.0%), and their mean IQ was The mildly retarded blacks had 78 study half-siblings, 20 of whom were retarded (25.6%). Their mean IQ was 79.5 (SD, 11.4). Thirteen of the mildly retarded black children had both full and half-siblings in the study. Two of the 16 full siblings were retarded; their mean IQ was 79.0 (SD, 11.9). Five of the 19 half-siblings were retarded; their mean IQ was 76.7 (SD, 11.9).

166 Nichols Cousins Cousins of the retarded children were identified following the mothers' report of any sisters or sisters-in-law registered in the NCPP.

6 166 Nichols Cousins Cousins of the retarded children were identified following the mothers' report of any sisters or sisters-in-law registered in the NCPP. The severely retarded white children had 23 first cousins in the study. None was retarded, and their mean IQ was (SD, 14.9). Four of 58 first cousins of the mildly retarded white children were retarded (6.9%), and the cousins' mean IQ was 92.1 (SD, 15.0). Like the full siblings of the retarded black children, the cousins of the severely and mildly retarded black children had similar scores. Four of 35 first cousins of the severely retarded blacks were retarded (11.4%), and the cousins' mean IQ was 86.5 (SD, 17.8). Twenty of 316 first cousins of the mildly retarded were retarded (6.3%), and their mean IQ was 87.3 (SD, 12.9). The cousins of both the severely and the mildly retarded had mean IQs similar to the NCPP black populations mean, whereas the full siblings of both retarded types had mean scores below that of the population. Twins The number of retarded twins in the cohort was too small for classical twins studies with respect to severe and mild retardation. Of 65 retarded twins in the NCPP, only 53 could be used for pairwise analyses because the cotwin had died or, in three cases, the zygosity had not been established. They were identified as monozygotic (MZ) or dizygotic (DZ) from the comparison of nine blood groups, finger and palm prints, and gross and microscopic examination of the placenta (Myrianthopoulos, 1970). The IQs of the 53 retarded twins are shown with the scores of their cotwins in Table II. Based on the IQ cutoff of 50, there was only one severely retarded MZ twin (IQ, 48); his IQ, however, was only two points lower than that of his cotwin, whose score fell just at the cutoff. For the DZ twins, an analysis similar to that performed for the siblings and cousins showed that five cotwins of severely retarded black children had a mean IQ of 79.2 (SD, 13.9), compared to 74.3 (SD, 17.6) for 23 cotwins of the mildly retarded. For the very small sample of white DZ pairs, the two cotwins of severely retarded had a mean IQ of 91.0, much higher than the mean of 66.6 for the five cotwins of mildly retarded. Overall, the cotwins of 12 of 18 retarded MZ twins and 14 of 35 DZ twins were retarded. Assessing risks to the cotwins is complicated by the fact that retardation in twins was more frequent than among singletons, by a factor of 2.7 in blacks and 1.9 in whites. There was generally a greater resemblance between MZ than DZ pairs, but because of the small samples, concordance for retardation was not significantly different between the

7 Familial Mental Retardation 167 Table II. IQs of Twin Pairs, One or Both Retarded, by Zygosity, Ethnicity, and Sex Monozygotic twins Dizygotic twins IQ Scores Sex IQ Scores Sex Black 48, 50 G 21, 88 BG 50, 54 G 25, 72 BG 51, 57 G 25, 83 BG 59, 69 B 34, 59 GB 60, 62 G 37, 94 BG 63, 83 G 52, 60 BG 65, 71 B 59, 73 BB 68, 82 B 60, 61 GB 68, 85 G 64, 67 BB 69, 73 G 64, 69 BB 69, 80 G 52, 80 BG 62, 120 BG 62, 88 BG 64, 73 BG 64, 74 BG 64, 83 BG 66, 95 GB 67, 73 GG 67, 74 BG 67, 99 GG 68, 73 GG 68, 82 BG 68, 92 GB White 60, 62 G 25, 94 BB 47, 88 BG 63, 69 GB 64, 67 GB 64, 70 GB two twin types. The interested reader may use the data presented for additional comparisons or to combine with other samples. The frequency of retardation and mean IQ among various types of relatives of the retarded children are summarized in Table III. As mentioned earlier, families with more than one affected child were not overrepresented in the NCPP population, so that for determining these frequencies, all affected individuals could be used as index cases. DISCUSSION These unique data from the NCPP provide an opportunity to compare familial patterns by severity of retardation and ethnic group. A possible criticism of familial studies from this project is that the children in the

168 Nichols Table IlL Frequency of Retardation and Mean IQ Among Relatives of Severely and Mildly Retarded Severely retarded Mildly retarded WISC IQ WlSC 1Q Relationship to Percentage Percentage

8 168 Nichols Table IlL Frequency of Retardation and Mean IQ Among Relatives of Severely and Mildly Retarded Severely retarded Mildly retarded WISC IQ WlSC 1Q Relationship to Percentage Percentage retarded N retarded Mean SD N retarded Mean SD White Monozygotic twin Dizygotic twin Full sibling I Half-sibling I 9.4 First cousin Black Monozygotic twin i Dizygotic twin Full sibling Half-sibling First cousin study are somehow a biased sample of all family members. There is, however, no evidence or reason to believe that the IQ distributions of study and nonstudy siblings differ. Unlike the controversial Roberts (1952) study, mild and severe retardation was determined on the basis of IQ scores (or, in a few cases, estimated IQ levels). The introduction of other criteria in the classification of cases, such as the presence of central nervous system pathology, might have increased the differences between the two groups, but IQs only were used to keep the criterion unambiguous and to see if patterns existed similar to the one Roberts was said to have found. Among whites, the data show a clear distinction between the relatives of the severely and the relatives of the mildly retarded children. This finding agrees with the purported results of Roberts that siblings of severely retarded children have higher IQs and less retardation than the siblings of mildly retarded children. Both the sibling and the first-cousin data for the white population suggest a qualitative difference between severe and mild retardation. Most cases of severe retardation are related to central nervous system pathology. There is much more speculation than evidence in the literature concerning the etiology of mild or culturalfamilial retardation (Zigler, 1978). In the current study, the 12-fold increase in the frequency of retardation among full siblings of mildly re-

Familial Mental Retardation 169 tarded whites (from 1.7% in the NCPP white population to 20.7%), compared with the 4-fold increase among first cousins (to 6.

9 Familial Mental Retardation 169 tarded whites (from 1.7% in the NCPP white population to 20.7%), compared with the 4-fold increase among first cousins (to 6.9%), could be consistent with genetic or environmental influences. The data for blacks lead to different conclusions. First, the family patterns do not show a clear distinction between the relatives of the severely and the relatives of the mildly retarded. The data support the hypothesis that the entire IQ distribution is shifted downward in the black population, so that "severely" retarded black children with IQs in the 40s are similar to the mildly retarded in terms of central nervous system pathology, socioeconomic status, and familial patterns. Only if the IQ cutoffs for defining severe and mild retardation are shifted downward are the distributions of siblings' IQs at all consistent with the two-group theory. None of the results of this study supports a genetic etiology of familial retardation in blacks. The increased risk of retardation to all types of relatives living with a retarded child--full siblings, half-siblings, and even the twins after considering the increased frequency of retardation in all twins (an increase which, of course, must be environmental)--was of the same order of magnitude, about four times the population frequency. On the other hand, first cousins, who lived apart from the retarded children, did not have increased frequencies of retardation or lower mean IQs. Among retardates with both full and half-siblings in the study, the mean IQ of the half-siblings was not higher than that of the full siblings, as would be expected under a polygenic hypothesis. Although the samples were small, there was not even a trend in this direction. In summary, the family data presented in this report support the traditional two-group theory of mental retardation in the white population. In the black population, where most retardation appeared to be of the cultural-familial type, environmental factors were implicated, a finding consistent with the notion that the downward shift in the IQ distribution also has an environmental basis. REFERENCES Akesson, H. O. (196t). Epidemiology and Genetics of Mental Deficiency in a Southern Swedish Population, Almquist and Wiksells, Uppsala. Broman, S. H., Nichols, P. L., and Kennedy, W. A. (1975). Preschool IQ: Prenatal and Early Developmental Correlates, Erlbanm, Hillsdale, N.J. Jensen, A. R. (1969). How much can we boost IQ and scholastic achievement? Harvard Educ. Rev. 39: Kamin, L. J. (1974). The Science and Politics oflq, Erlbaum, Potomac, Md. Myrianthopoulos, N. C. (1970). An epidemiologic survey of twins in a large, prospectively studied population. Am. J. Hum. Genet. 22:

170 Nichols Myrianthopoulos, N. C., and French, K. S. (1968). An application of the U.S. Bureau of the Census socioeconomic index to a large, diversified patient population. Soc. Sci. Med. 2:283-299.

10 170 Nichols Myrianthopoulos, N. C., and French, K. S. (1968). An application of the U.S. Bureau of the Census socioeconomic index to a large, diversified patient population. Soc. Sci. Med. 2: Penrose, L. S. (1939). Intelligence test scores of mentally defective patients and their relatives. Br. J. Psychol. 30:1-18. Roberts, J. A. F. (1940). Studies on a child population. V. The resemblance in intelligence between sibs. Ann. Eugen. Lond. 10" Roberts, J. A. F. (1952). The genetics of mental deficiency. Eugen. Rev. 44: Vandenberg, S. G. (1968). The nature and nurture of intelligence. In Glass, D. C. (ed.), Genetics, Rockefeller University Press and The Russell Sage Foundation, New York. Zigler, E. (1967). Familial mental retardation: A continuing dilemma. Science 155: Zigler, E. (1978). National crisis in mental retardation research. Am. J. Ment. Def 83"1- " Edited by Thomas J. Bouchard, Jr.

Chapter 2 Interactions Between Socioeconomic Status and Components of Variation in Cognitive Ability

Chapter 2 Interactions Between Socioeconomic Status and Components of Variation in Cognitive Ability Eric Turkheimer and Erin E. Horn In 3, our lab published a paper demonstrating that the heritability