RESEARCH
Minor Neurological Dysfunction Is More Closely Related to Learning Difficulties than to Behavioral Problems Mijna Hadders-Algra and Bert C.L. Touwen
In a group of 570 nine-year-old children (315 boys and 255 girls) without an overtly handicapping neurological condition, relationships were studied between the presence of minor neurological dysfunction (MND) on the one hand and cognitive and behavioral problems on the other hand. The aim of the study was to investigate whether MND was more closely related to learning problems than to behavioral difficulties. The group was a subpopulation of the birth cohort of the Groningen Perinatal Project in the Netherlands, in which perinatal developmental relationships are studied. The age-specific and standardized neurological examination technique of Touwen (1979) was used, resulting in a description of the neurological condition in terms of the number of deviant clusters. The absence of deviant clusters indicated a normal neurological condition (n=418); children with one or two deviant clusters were classified as MND-1 (n-86) and those with more than two deviant clusters as MND-2 (n=66). The presence and severity of MND was significantly related to poor performance on standardized reading spelling and arithmetic tests. MND was also related to ratings of distracttble behavior on parent and teacher questionnaires, but to a lesser extent than the cognitive problems. Ratings of so-called "troublesome" and "timid" behavior were not related to MND. Our conclusion is that learning problems are more closely related to MND than are behavioral difficulties. This has implications for prevention and intervention: In the former the focus should be more on biological hazards, in the latter on environment and rearing attitudes.
I
n previous papers we reported significant relationships between the presence of minor neurological dysfunction (MND) and learning and behavioral problems in 6- and 9year-old children belonging to a 3-year cohort of children born in the Groningen University Hospital (HaddersAlgra, Huisjes, & Touwen, 1988a, 1988c; Hadders-Algra, Touwen, & Huisjes, 1986; Hadders-Algra, Touwen, Olinga, & Huisjes, 1985). These findings are in line with other reports indicating a higher frequency of neurological "soft signs" in children with learning problems (Adams, Kocsis, &
Estes, 1974; Berninger & Colwell, 1985; Rie, 1987; Shaffer, O'Connor, Shafer, & Prupis, 1984) and in those with hyperactivity and/or an attention deficit (Reeves & Werry, 1987; Shaffer et al., 1984; Stine, Saratsiotis, & Mosser, 1975; Taylor, 1986). Controversy exists over whether cognitive difficulties are more closely related to minor neurological dysfunction than are behavioral problems. The large majority of studies on this subject concern reports on children referred for learning disabilities or behavior disorders. The reported incidences of soft signs vary considerably (e.g.,
Deuel & Robinson, 1987), which can be explained largely by differences in subject selection (target symptom, age, sex) and examination method. It is difficult to decide on the basis of these scattered data whether learning problems are more closely related to MND than are behavioral difficulties. Studies of children not selected on the basis of learning or behavioral problems provide conflicting data. Rutter (1981, 1982) concluded, on the basis of a review of studies on psychological sequelae of brain damage in children, that the relation between brain damage and cognitive impairment was stronger than the relation between brain injury and psychiatric disorder. In contrast, Nichols and Chen (1981) and Herzig (1987), who followed children from birth, reported similar frequencies of neurological dysfunction in children with learning difficulties and those with behavioral problems. A difference in the extent to which cognitive problems and behavioral difficulties are related to organically based minor dysfunction of the central nervous system (CNS) has important implications for the prevention and treatment of these problems. Therefore, the aim of the present article is to investigate whether a difference exists in the magnitude of correlation between MND and learning problems on the
JOURNAL OF LEARNING DISABILITIES Downloaded from ldx.sagepub.com at The University of Iowa Libraries on June 9, 2015 VOLUME 25, NUMBER 10, DECEMBER 1992
PAGES 649-657
650
JOURNAL OF LEARNING DISABILITIES
one hand, and MND and behavioral difficulties on the other hand, in groups of children without a referral bias. The 9-year follow-up data of the birth cohort of the Groningen Perinatal Project (GPP) in the Netherlands, which include detailed neurological information as well as data on learning and behavioral problems, allow such an analysis.
Method Subjects The study group is part of the GPP, a cohort study of perinatal relationships in children born from 1975 to
1978 in the Groningen University Hospital. All children (N=3,162) were examined neurologically in their newborn period according to Prechtl's (1977) standardized method (see Huisjes et al., 1980; Jurgens-van der Zee et al., 1979). We wanted to study obstetrical-neonatal-developmental relationships in the children at age 9. Therefore, all neonatally neurologically deviant (A) infants (n = 160) and a random sample of nondisabled (N) (n = 322) and mildly abnormal (S) (n=322) newborns were invited to our laboratory for follow-up. From the latter two groups, 70 and 86 children, respectively, had been examined at the age of 6 years; 57 children (22 neonatally N, 22 S, and 13 A) were lost to follow-
up. Children with overt handicaps (n=21) were excluded from the present analysis. At 9 years an age-adequate and standardized neurological examination (Touwen, 1979) was carried out, in which special attention was paid to the presence of minor neurological dysfunction. The children were classified as nondisabled, MND, or abnormal. Minor neurological dysfunction consists of neurological deviations that do not lead to an overt handicap, such as the presence of choreiform dyskinesia, mild hypotonia, mild coordination problems, and slight deviations of fine manipulative ability. The classification abnormal implied the presence of a handicapping neurological condition,
TABLE 1 Cluster Profile of the Neurological Examination According to Touwen (1979) Cluster
Compound of:
Criteria for the presence of deviant cluster
1. Posture and muscle tone
Posture during sitting, standing, wa Muscle tone
Postural deviations, such as collapse, asymmetries, hyperextension Consistent changes of muscle tone
2. Reflexes
Biceps Knee Ankle Foot-sole response
Increased or decreased intensities/thresholds Asymmetries Babinski sign
3. Coordination and balance
Finger-nose test Fingertip-touching test Diadochokinesis Kicking Knee-heel test Reaction to push (sitting, standing) Romberg Tandem gait Standing on one leg
Two or more tests inappropriate for age
4. Fine manipulative ability
Finger-opposition test, smoothness Finger-opposition test, transition
Two or more tests inappropriate for age
Follow-a-finger test Circle test 5. Choreiform dyskinesia
During spontaneous motility Test with extended arms Face, eyes, tongue
Marked choreiform movements of distal and facial muscles Slight or marked choreiform movements of proximal muscles, eyes, or tongue
6. Rarely miscellaneous dysfunctions
Motility of facial musculature Position and movements of eyes Associated movements during diadochokinesis, finger-opposition test, walking on toes and heels
Sixth and seventh nerve palsy Excessive amount of associated movements (for
Note. From Hadders-Algra, Huisjes, and Touwen (1988b). Downloaded from ldx.sagepub.com at The University of Iowa Libraries on June 9, 2015
651
VOLUME 25, NUMBER 10, DECEMBER 1992
such as cerebral palsy, but these children were excluded, as it could be expected that they would differ behaviorally and with respect to cognitive development from the nonhandicapped children. This leaves a group of 570 clinically normal children—the same subjects described in HaddersAlgra, Huisjes, and Touwen (1988c). The children with MND were classified on the basis of a cluster profile summarizing the neurological findings (Touwen, 1979). The scoring criteria for the clusters, which consist of functional subsystems of the CNS, are shown in Table 1. Repeated, independent examinations of a random sample of the group by both authors revealed that the cluster scoring was consistent. Children were classified as MND-2 if more than two clusters were scored deviant, and MND-1 if there were one or two deviant clusters. The neurological results of the 9-year examination are shown in Table 2. Reading, spelling, and arithmetic were assessed with short, standardized Dutch tests (Brus & Voeten, 1972; Mommers, 1967; Ojemann, 1973, respectively). These tests are commonly used in the Netherlands for the evaluation of educational achievement. Children scoring in the lowest decile were regarded as "poor performers." Parents and teachers supplied information on school achievement and behavior. For the latter we used a questionnaire with 13 "balanced" questions, each item having three alternatives: "All children are sometimes. . . . Do you regard your/this child as (a) more . . . or (b) less . . . or (c) don't you see any difference." Factor analysis of 12 of the behavioral items revealed the presence of three factors: distractible, troublesome, and timid behavior (see Table 3). The highest decile was taken as a cutting point for undesirable behavior. The factor distractible is in accordance with the description of attention deficithyperactivity disorder in DSM-III-R (American Psychiatric Association, 1987). One question dealt with clumsiness; the answers were evaluated separately. School failure was defined as placement in an inappropriate grade
for age (which is an indicator of poor school achievement in the Netherlands) or attending a school for special education, which in the Netherlands can imply a school for children with low intelligence or children of normal intelligence but suffering from specific learning problems. IQ measures were not available. However, the fact that, with very few exceptions, most children went either to a regular school or to a school for special education where normal intelligence was a prerequisite indicates that the number of children with an IQ below 80 was negligible. Chi-square and Fisher's exact tests were used for statistical analysis.
quiring normal intelligence for the nondisabled children and for the MND-1 and MND-2 children is presented in Figure 1: Half of the children needing special education belong to the small part of the study population classified as MND-2. The difference among the three groups is statistically significant. Reading, spelling, and arithmetic achievements also showed a statistically significant relationship with the neurological classification (see Table 4). Spelling problems were about twice as frequent as reading difficulties (n = 119 and n=64, respectively). The majority (88%) of children with reading problems also showed spelling problems, a commonly reported finding (see Frith, 1984). The presence of both reading and spelling (R+S) problems was related to the neurological condition (see Table 4). About half of the children
Results The frequency of children's attending schools for special education re-
TABLE 2 Neurological Classification iri Boys and Girls Neurological classification Nondisabled MND-1 MND-2 Total
Boys
Girls
Total
Sex ratio
213 57 45 315
205 29 21 255
418 86 66 570
1.0:1 2.0:1 2.1:1
TABLE 3 Factor Analysis Behavior: Varimax Loadings of 0.30 or Greater Explained variance
(%)
Items of questionnaire Concentration Distractibility Attention span Adaptability Irritable Hyperactive Easy to handle Headstrong Temper tantrums Easily frightened Shy, withdrawn Anxious
Distractible P (59) T(60)
•0.74 0.78 -0.62
-0.85 0.87 -0.77 -0.40 0.32
Troublesome P(27) T(21)
Timid P(17) T(16)
-0.34 0.62 0.51 -0.58 0.55 0.50
-0.33
-0.40 0.72 0.57 -0.73 0.72 0.46
Note. P = parents; T = teachers; from Hadders-Algra, Huisjes, and Touwen (1988c).
Downloaded from ldx.sagepub.com at The University of Iowa Libraries on June 9, 2015
0.48 0.33 0.55
0.50 0.44 0.66
652
JOURNAL OF LEARNING DISABILITIES
80-1
7CH
~| Normal school 6CH
5CH
40H
30H
for VZZA sSchool pecial education
attention problems. In the group of neurologically normal children, only one child (0.2%, a boy) showed this triad of problems, which is significantly less than in the MND group (Fisher p
Minor Neurological Dysfunction Is More Closely Related to Learning Difficulties than to Behavioral Problems Mijna Hadders-Algra and Bert C.L. Touwen
In a group of 570 nine-year-old children (315 boys and 255 girls) without an overtly handicapping neurological condition, relationships were studied between the presence of minor neurological dysfunction (MND) on the one hand and cognitive and behavioral problems on the other hand. The aim of the study was to investigate whether MND was more closely related to learning problems than to behavioral difficulties. The group was a subpopulation of the birth cohort of the Groningen Perinatal Project in the Netherlands, in which perinatal developmental relationships are studied. The age-specific and standardized neurological examination technique of Touwen (1979) was used, resulting in a description of the neurological condition in terms of the number of deviant clusters. The absence of deviant clusters indicated a normal neurological condition (n=418); children with one or two deviant clusters were classified as MND-1 (n-86) and those with more than two deviant clusters as MND-2 (n=66). The presence and severity of MND was significantly related to poor performance on standardized reading spelling and arithmetic tests. MND was also related to ratings of distracttble behavior on parent and teacher questionnaires, but to a lesser extent than the cognitive problems. Ratings of so-called "troublesome" and "timid" behavior were not related to MND. Our conclusion is that learning problems are more closely related to MND than are behavioral difficulties. This has implications for prevention and intervention: In the former the focus should be more on biological hazards, in the latter on environment and rearing attitudes.
I
n previous papers we reported significant relationships between the presence of minor neurological dysfunction (MND) and learning and behavioral problems in 6- and 9year-old children belonging to a 3-year cohort of children born in the Groningen University Hospital (HaddersAlgra, Huisjes, & Touwen, 1988a, 1988c; Hadders-Algra, Touwen, & Huisjes, 1986; Hadders-Algra, Touwen, Olinga, & Huisjes, 1985). These findings are in line with other reports indicating a higher frequency of neurological "soft signs" in children with learning problems (Adams, Kocsis, &
Estes, 1974; Berninger & Colwell, 1985; Rie, 1987; Shaffer, O'Connor, Shafer, & Prupis, 1984) and in those with hyperactivity and/or an attention deficit (Reeves & Werry, 1987; Shaffer et al., 1984; Stine, Saratsiotis, & Mosser, 1975; Taylor, 1986). Controversy exists over whether cognitive difficulties are more closely related to minor neurological dysfunction than are behavioral problems. The large majority of studies on this subject concern reports on children referred for learning disabilities or behavior disorders. The reported incidences of soft signs vary considerably (e.g.,
Deuel & Robinson, 1987), which can be explained largely by differences in subject selection (target symptom, age, sex) and examination method. It is difficult to decide on the basis of these scattered data whether learning problems are more closely related to MND than are behavioral difficulties. Studies of children not selected on the basis of learning or behavioral problems provide conflicting data. Rutter (1981, 1982) concluded, on the basis of a review of studies on psychological sequelae of brain damage in children, that the relation between brain damage and cognitive impairment was stronger than the relation between brain injury and psychiatric disorder. In contrast, Nichols and Chen (1981) and Herzig (1987), who followed children from birth, reported similar frequencies of neurological dysfunction in children with learning difficulties and those with behavioral problems. A difference in the extent to which cognitive problems and behavioral difficulties are related to organically based minor dysfunction of the central nervous system (CNS) has important implications for the prevention and treatment of these problems. Therefore, the aim of the present article is to investigate whether a difference exists in the magnitude of correlation between MND and learning problems on the
JOURNAL OF LEARNING DISABILITIES Downloaded from ldx.sagepub.com at The University of Iowa Libraries on June 9, 2015 VOLUME 25, NUMBER 10, DECEMBER 1992
PAGES 649-657
650
JOURNAL OF LEARNING DISABILITIES
one hand, and MND and behavioral difficulties on the other hand, in groups of children without a referral bias. The 9-year follow-up data of the birth cohort of the Groningen Perinatal Project (GPP) in the Netherlands, which include detailed neurological information as well as data on learning and behavioral problems, allow such an analysis.
Method Subjects The study group is part of the GPP, a cohort study of perinatal relationships in children born from 1975 to
1978 in the Groningen University Hospital. All children (N=3,162) were examined neurologically in their newborn period according to Prechtl's (1977) standardized method (see Huisjes et al., 1980; Jurgens-van der Zee et al., 1979). We wanted to study obstetrical-neonatal-developmental relationships in the children at age 9. Therefore, all neonatally neurologically deviant (A) infants (n = 160) and a random sample of nondisabled (N) (n = 322) and mildly abnormal (S) (n=322) newborns were invited to our laboratory for follow-up. From the latter two groups, 70 and 86 children, respectively, had been examined at the age of 6 years; 57 children (22 neonatally N, 22 S, and 13 A) were lost to follow-
up. Children with overt handicaps (n=21) were excluded from the present analysis. At 9 years an age-adequate and standardized neurological examination (Touwen, 1979) was carried out, in which special attention was paid to the presence of minor neurological dysfunction. The children were classified as nondisabled, MND, or abnormal. Minor neurological dysfunction consists of neurological deviations that do not lead to an overt handicap, such as the presence of choreiform dyskinesia, mild hypotonia, mild coordination problems, and slight deviations of fine manipulative ability. The classification abnormal implied the presence of a handicapping neurological condition,
TABLE 1 Cluster Profile of the Neurological Examination According to Touwen (1979) Cluster
Compound of:
Criteria for the presence of deviant cluster
1. Posture and muscle tone
Posture during sitting, standing, wa Muscle tone
Postural deviations, such as collapse, asymmetries, hyperextension Consistent changes of muscle tone
2. Reflexes
Biceps Knee Ankle Foot-sole response
Increased or decreased intensities/thresholds Asymmetries Babinski sign
3. Coordination and balance
Finger-nose test Fingertip-touching test Diadochokinesis Kicking Knee-heel test Reaction to push (sitting, standing) Romberg Tandem gait Standing on one leg
Two or more tests inappropriate for age
4. Fine manipulative ability
Finger-opposition test, smoothness Finger-opposition test, transition
Two or more tests inappropriate for age
Follow-a-finger test Circle test 5. Choreiform dyskinesia
During spontaneous motility Test with extended arms Face, eyes, tongue
Marked choreiform movements of distal and facial muscles Slight or marked choreiform movements of proximal muscles, eyes, or tongue
6. Rarely miscellaneous dysfunctions
Motility of facial musculature Position and movements of eyes Associated movements during diadochokinesis, finger-opposition test, walking on toes and heels
Sixth and seventh nerve palsy Excessive amount of associated movements (for
Note. From Hadders-Algra, Huisjes, and Touwen (1988b). Downloaded from ldx.sagepub.com at The University of Iowa Libraries on June 9, 2015
651
VOLUME 25, NUMBER 10, DECEMBER 1992
such as cerebral palsy, but these children were excluded, as it could be expected that they would differ behaviorally and with respect to cognitive development from the nonhandicapped children. This leaves a group of 570 clinically normal children—the same subjects described in HaddersAlgra, Huisjes, and Touwen (1988c). The children with MND were classified on the basis of a cluster profile summarizing the neurological findings (Touwen, 1979). The scoring criteria for the clusters, which consist of functional subsystems of the CNS, are shown in Table 1. Repeated, independent examinations of a random sample of the group by both authors revealed that the cluster scoring was consistent. Children were classified as MND-2 if more than two clusters were scored deviant, and MND-1 if there were one or two deviant clusters. The neurological results of the 9-year examination are shown in Table 2. Reading, spelling, and arithmetic were assessed with short, standardized Dutch tests (Brus & Voeten, 1972; Mommers, 1967; Ojemann, 1973, respectively). These tests are commonly used in the Netherlands for the evaluation of educational achievement. Children scoring in the lowest decile were regarded as "poor performers." Parents and teachers supplied information on school achievement and behavior. For the latter we used a questionnaire with 13 "balanced" questions, each item having three alternatives: "All children are sometimes. . . . Do you regard your/this child as (a) more . . . or (b) less . . . or (c) don't you see any difference." Factor analysis of 12 of the behavioral items revealed the presence of three factors: distractible, troublesome, and timid behavior (see Table 3). The highest decile was taken as a cutting point for undesirable behavior. The factor distractible is in accordance with the description of attention deficithyperactivity disorder in DSM-III-R (American Psychiatric Association, 1987). One question dealt with clumsiness; the answers were evaluated separately. School failure was defined as placement in an inappropriate grade
for age (which is an indicator of poor school achievement in the Netherlands) or attending a school for special education, which in the Netherlands can imply a school for children with low intelligence or children of normal intelligence but suffering from specific learning problems. IQ measures were not available. However, the fact that, with very few exceptions, most children went either to a regular school or to a school for special education where normal intelligence was a prerequisite indicates that the number of children with an IQ below 80 was negligible. Chi-square and Fisher's exact tests were used for statistical analysis.
quiring normal intelligence for the nondisabled children and for the MND-1 and MND-2 children is presented in Figure 1: Half of the children needing special education belong to the small part of the study population classified as MND-2. The difference among the three groups is statistically significant. Reading, spelling, and arithmetic achievements also showed a statistically significant relationship with the neurological classification (see Table 4). Spelling problems were about twice as frequent as reading difficulties (n = 119 and n=64, respectively). The majority (88%) of children with reading problems also showed spelling problems, a commonly reported finding (see Frith, 1984). The presence of both reading and spelling (R+S) problems was related to the neurological condition (see Table 4). About half of the children
Results The frequency of children's attending schools for special education re-
TABLE 2 Neurological Classification iri Boys and Girls Neurological classification Nondisabled MND-1 MND-2 Total
Boys
Girls
Total
Sex ratio
213 57 45 315
205 29 21 255
418 86 66 570
1.0:1 2.0:1 2.1:1
TABLE 3 Factor Analysis Behavior: Varimax Loadings of 0.30 or Greater Explained variance
(%)
Items of questionnaire Concentration Distractibility Attention span Adaptability Irritable Hyperactive Easy to handle Headstrong Temper tantrums Easily frightened Shy, withdrawn Anxious
Distractible P (59) T(60)
•0.74 0.78 -0.62
-0.85 0.87 -0.77 -0.40 0.32
Troublesome P(27) T(21)
Timid P(17) T(16)
-0.34 0.62 0.51 -0.58 0.55 0.50
-0.33
-0.40 0.72 0.57 -0.73 0.72 0.46
Note. P = parents; T = teachers; from Hadders-Algra, Huisjes, and Touwen (1988c).
Downloaded from ldx.sagepub.com at The University of Iowa Libraries on June 9, 2015
0.48 0.33 0.55
0.50 0.44 0.66
652
JOURNAL OF LEARNING DISABILITIES
80-1
7CH
~| Normal school 6CH
5CH
40H
30H
for VZZA sSchool pecial education
attention problems. In the group of neurologically normal children, only one child (0.2%, a boy) showed this triad of problems, which is significantly less than in the MND group (Fisher p
