CLASSIFICATION

OF MENTAL RETARDATION

S u p p l e m e n t t o T H E AMERICAN JOURNAL OF PSYCHIATRY

Vol. 128, N o . 11, M a y 1972

CONTENTS

Mental Retardation: Development of an International

Classification Scheme Michael J. Begab and Gerald D. La Veck

Fifth WHO Seminar on Psychiatric Diagnosis, Classification,

and Statistics

Some Thoughts on the Classification of Mental Retardation

in the United States of America George Tarjan and Leon Eisenberg

Differing Concepts of Diagnosis as a Problem in Classification

Jack R. Ewalt

Comments on the ICD Classification of Mental Retardation

Joseph Wortis

A Note on the International Statistical Classification of

Mental Retardation Jack Tizard

The Problem of the Classification of Mental Retardation

G.E. Suhareva

Classification and Mental Retardation: Issues Arising in the

Fifth WHO Seminar on Psychiatric Diagnosis, Classification,
and Statistics G. Tarjan, J. Tizard, M. Rutter, M. Begab,
E.M. Brooke, F. de la Cruz, T.-y. Lin, H. Montenegro,
H. Strotzka, and N. Sartorius

Publication of this supplement was supported by contract NIH72-C-563 from the National Institute of Child Health and
Human Development, National Institutes of Health.

Copyright 1972 American Psychiatric Association

SUPPLEMENT

Classification of Mental Retardation

Mental Retardation: Development of an

International Classification Scheme

ENTAL RETARDATION is a universal phenomenon. It occurs, with

varying degrees of frequency, in families from all walks of life in
both developed and underdeveloped countries. In its most severe forms
it is a source of great trauma, hardship, and despair to parents and is
an economic and social burden to communities. Even the milder forms
of intellectual handicap pose serious threats to individual self-fulfillment, family security, and national productivity. The most affluent of
nations can ill afford such losses in their human resources.
The prevention and treatment of mental retardation on both the
individual and societal levels rest fundamentally on a fuller understanding of its causes and pathogenesis, on concerned and skilled professional practitioners, and on the commitment of appropriate resources at all levels of government. Efforts to reach the first of these
goalswhich is essential to classificationhave been greatly enhanced
in recent decades through basic and applied research. During this period we have identified additional clinical syndromes, developed a technology for prenatal diagnosis and prevention, improved nutritional and
medical intervention techniques, and made progress toward solving the
mysteries surrounding the transmission of genetic materials at the cellular level.
In the behavioral sciences much has been learned about the impact of
environmental deprivation on mental growth and the compensating effects of early stimulation, about methods for promoting language
development and reading skills, and about the untapped capacities of
many retarded individuals for socially useful living. Perhaps most important of all is the growing recognition that in most forms of retardation, even where a single etiological factor can be isolated, the
individual's functional performance is the product of the interaction of
his biological makeup and environmental events and can be modified.
The potential for behavioral change, sometimes to the point of reversibility, represents one of the most significant concepts in the field to
emerge in recent years.

The changing attitudes of psychiatrists, pediatricians, and obstetricians toward the mentally retarded stem in part from this new conceptualization of the problem and the growing conviction that even where
"cures" are not possible, informed treatment of the individual and his
family can significantly aid life adjustment. To capitalize more fully on
this burgeoning interest, these disciplines need more precise information on hazards to fetal development, symptomatology and treatment
potentials for specific diagnostic conditions, and the values and limitations of psychological test measurements. Furthermore, to keep abreast
of new discoveries and program developments, these disciplines must
share a terminology and language that permit communication. Our
failure in this latter area has seriously handicapped efforts of professionals from different countries to learn from one another.
The World Health Organization, mindful of these deficiencies and of
our increasing fund of knowledge, has embarked upon a series of
seminars to develop an international scheme for the diagnosis, classification, and reporting of statistics in psychiatric disorders, including
mental retardation. This effort comes at a most opportune time. Comparative data among countries on the incidence and prevalence of
mental retardation and the factors with which specific conditions are
associated are not highly reliable. Although there are significant variations in prenatal care, population homogeneity, disease control, degree
of environmental deprivation, and other factors causative or contributory to mental retardation, reported statistical differences may be more
artifactual than real. Differences in the definition and conceptualization
of mental retardation, inadequacies and variations in classification
schemes used, confusion of terminology, and cultural variability in
demands and expectations for human performance are only a few of the
artifacts that preclude valid comparisons. Within and among countries,
meaningful planning for the retarded cannot be accomplished until
these issues are resolved.
The 1969 seminar, cosponsored by the World Health Organization
and the National Institute of Child Health and Human Development,
was a milestone in the realization of these goals. It is clear that the
complex issues confronted will require continuing attention, but meaningful dialogue has begun and a sounder base for assessing the extent
and diversity of this problem is being established. Community planners
and professional practitioners should profit from this activity, but the
ultimate beneficiaries and the raison d'etre of the seminar will be the
mentally retarded and their families.
MICHAEL J. BEGAB, P H . D .
GERALD D. LAVECK, M.D.

Fifth WHO Seminar on Psychiatric Diagnosis,

Classification, and Statistics
This report describes in detail the problems
of adequately classifying mental retardation
and the recommendations made for the
forthcoming ninth revision of the International Classification of Diseases. In general
the Seminar agreed that the current classification of mental retardation is inadequate
and that a multiaxial scheme should be
adopted. This scheme would consist of three
axes: 1) intellectual level, 2) associated or
etiological factors, and 3) clinical psychiatric
system, and would require that each axis be
recorded.
Information that should be
considered in classifying mental retardation
includes: degree, organic aspects, psychiatric
and behavioral aspects, and psychosocial aspects. The participants also considered the
need to develop a glossary and how best to
promote the effective use of the ICD.

was the fifth in the World

Health Organization's (WHO) ten-year
program on "Psychiatric Diagnosis, Classification, and Statistics." Previous seminars'
dealt with "Problems of Functional Psychoses, Particularly Schizophrenia" ( 2 ) ,
"Reactive Psychoses" (3), "Mental Disorders
in Children" (4), and "Psychiatric Disorders
of Old Age" (5). In the development of a
classification of mental disorders, four prinHIS SEMINAR

ciples had been previously adopted.

1. The study of the process of psychiatric
diagnosis, as provided by the diagnostic exercise, should be used as the basis for understanding the different schemes that psychiatrists of different schools employ. Major
sources of variation and error should be
identified in order to improve the reliability
and validity of psychiatric diagnosis.
.2. In view of the variety of theoretical
concepts regarding the etiology and pathogenesis of mental disorders, and because
of the paucity of evidence that might lead to a
choice between theories, emphasis should be
placed on the use of solid clinical facts as a
starting point in developing a classification.
3. The definition of terms should be
operational and capable of clinical application.
4. Any scheme agreed upon at a seminar
will be tested through national and international exercises and through further refinement and revision before a final recommendation is made for the revision of
the International Classification of Diseases
(ICD) in 1975.
The following activities that were carried
out in different countries since the last seminar were reported by the participants (see
appendix 1): 1) Diagnostic exercises using
case histories from previous seminars were
carried out in Bulgaria, Czechoslovakia, Japan, and the U.S.S.R.; 2) glossaries and
diagnostic manuals were prepared in a number of countries that were represented at the
Seminar; 3) trial use of the Paris classification of children's disorders is being planned
in France, the United Kingdom, and the
United States. The importance of such experiments was particularly stressed by many
participants as being essential for the improvement of the ICD.
Diagnostic Exercises
The procedure, which had been successfully
used in previous seminars, of beginning with a

detailed discussion of problems arising from

the analysis of case histories and of videotapes, was again used in this one. Here the
purpose was to enable participants first, to
study a common set of case materials relating
to mental retardation and to record their
judgments about it; second, to discuss disagreements, ambiguities, and points of
controversy; and third, through the elucidation of specific problems to approach general
principles of diagnosis, classification, and
statistics in this area.
Case History Exercise
Several months before the meeting participants received case histories of 11 patients
that they were asked to read and to analyze.
The cases were chosen to illustrate problems
that were related to difficulties in diagnosis
regarding level of intellectual retardation, the
causes of the patients' retardation, the diagnosis of psychiatric disorder accompanying
the mental retardation, and the differential
diagnosis of mental retardation and developmental disorders. The patients included
some who were typical and others who were
borderline or had an uncertain diagnosis that
posed considerable problems. Each case history followed a standard format, giving the
reasons for the patient's admission, his family
history, information about his siblings, his
developmental history, his history of neurological and physical illnesses, the results of
current physical examination, neurological
examination, laboratory findings, and psychological evaluation, and some information
about his hospital course or disposal. Participants were invited to make a diagnosis of
the patient, using their own terminology, to
indicate whether the diagnosis was firm or
provisional, and to code their diagnosis according to the ICD rubrics by using one or
more than one category. They were given
standardized forms on which to record their
judgments; these were then returned to WHO
for analysis. Summaries of the data indicating diagnostic agreement and disagreement
on the patients, as well as the comparison of
diagnoses, were handed out at the meeting
and participants brought with them the case
material they had been sent. The excellent
preparation of case histories by the U.S. colleagues in the preparatory committee and in
the National Institute of Child Health and
Human Development (NICHD) was widely

appreciated, and there was general comment

throughout the meeting on the outstandingly
good choice of cases that highlighted the chief
issues to be considered during the discussions.
Videotape Diagnostic Exercise
Three patients who were representative of
different types of problems were shown on
videotape to the participants. The purpose of
this exercise was, again, to illustrate different
facets of the diagnostic problem by showing a
videotaped interview. The three patients
shown illustrated problems of the differential
diagnosis of developmental disorders and
mental retardation, the problem of the differential diagnosis of child psychosis and
mental retardation, and a diagnostic problem
that was complicated by the fact that one of
the patients was an identical twin. Following
a recommendation made at a past conference, the technique adopted at the Seminar
was first to show the videotape to the participants, who were asked to rate the patient
according to his behavioral characteristics,
knowing only his sex and age. Participants
were then given a case history and were
invited to ask for additional information that
was not included in it. They then made a final
diagnosis of the patient, rating this as either
firm or provisional.
The method of presentation of the video
material proved to be reasonably satisfactory
in that it enabled analyses to be made of ratings of observed behavior and of this together
with the written case material. It was
thought, however, that the discussion would
have been more fruitful had the videotapes
been available for replay during the discussion, and it was recommended that arrangements be made for this in future seminars.
It was also recommended that for future seminars there be testing and revision of the
evaluation sheets before the meeting.
Principal Topics of Discussion
Agreement on Diagnosis and Classification
Level of intellectual retardation. For each
of the patients in the case history exercise, the
participants were asked to categorize the
level of mental retardation according to the
six-point coding (310-315) given in the eighth
revision of ICD (ICD-8) (6). In most cases
this caused no major difficulties, and there
was a high level of agreement80 to 90 per-

cent of the participants giving the same coding for the level of retardation. The only
patient over whom there was appreciable disagreement was a six-week-old boy with
chromosomal abnormalities. It was agreed
that it was not possible to make a meaningful
assessment of intelligence in infancy and that
judgments on level of retardation in very
young children could only be approximate.
Participants were unanimous in agreeing that
the greatest difficulties in assessing retardation existed for individuals from minority
groups, from cultures other than those in
which the participants were living, and from
socially deprived communities. Whereas
intelligence tests are of great value in assessing level of retardation, they should never be
used in isolation from clinical considerations
of social-adaptive functioning. When
individuals come from cultures other than
those used for the standardization of intelligence quotient (IQ) tests, the tests are of
limited value.

overlaps other parts of ICD-8, and there is no

clear instruction as to whether the same
condition should be recorded under one or
both headings. Everyone agreed that it was
most important to record the biological disorders underlying mental retardation, but
dissatisfaction was expressed with the fourthdigit coding in that it necessitated judgments
about hypothetical causes (which were shown
to be unreliable) rather than recording associated neurological handicaps. This issue was
returned to later in the Seminar.
Psychiatric or behavioral disorder accompanying mental retardation. Several of the
patients showed disorders of behavior, as well
as mental retardation. For example, the third
patient in the case history exercise was a hostile, extremely hyperactive child who had
started several fires and who had gotten into
frequent fights. However, only ten of the 16
participants made a diagnosis outside of the
mental retardation section and even these ten
coded psychiatric disorder under three difCausation of retardation. The fourth digit ferent headings. The same issue arose in
in the ICD-8 mental retardation coding, that connection with one of the patients shown in
which specifies the cause of the disorder, gave the videotape exercise. Most of the particirise to no difficulties in patients who had pants diagnosed psychosis of some type, but
a clearly defined disease or disorder. Thus, this was coded in several different ways.
cases of phenylketonuria and chromosomal Some coded 295.8 ("childhood schizoa b n o r m a l i t y r e s u l t e d in u n a n i m o u s phrenia") without recording anything in the
agreement on fourth-digit codings of .2 and mental retardation section; some coded 310.7
("mental retardation following major psy.5, respectively.
However, in patients who did not have a chiatric disorder") without coding the type of
definite brain dysfunction or who did have disorder separately; some diagnosed psysome probable or definite organic brain dis- chosis but coded it under 308 ("behavior disorder, but one that did not fit the criteria for orders of childhood"); and some diagnosed
any specific disease, agreement on fourth- psychosis but coded only mental retardation.
digit coding was extremely poor. This was
It was generally agreed that this state of
due to uncertainties about etiology and to the affairs was most unsatisfactory. In spite of a
problems of multiple factors in etiology; thus high level of agreement on diagnosis, there
different participants coded different etio- was a very low level of agreement on coding.
logic factors. For example, in case one, a child This arose through three factors: 1) no satiswith a definite neurological disorder of an factory means of coding child psychiatric or
unknown type that had been present since behavior disorders; 2) no instruction on how
birth, codings of .1, .2, .3, .4, .6, and .9 were many diagnoses to code when a psychiatric
all used, and one participant made a diagno- disorder accompanies mental retardation;
sis under Section VI (Diseases of the Nervous and 3) unfamiliarity with sections of ICD-8
System and Sense Organs) rather than under outside of Section V. This issue was further
Section V (Mental Disorders).
discussed in relation to the report of the Paris
This patient and several others caused two Seminar.
important points to be raised: 1) Most cases
Developmental disorder. Several of the
of mental retardation, even those due to or- patients in both the case history and in the
ganic brain dysfunction, are not associated videotape exercises showed a severe delay in
with clearly diagnosable diseases, and 2) the the development of speech and/or language
fourth digit of the mental retardation coding that could not be accounted for in terms of

mental retardation. For example, the second

patient in the videotape exercise was a fiveyear-old boy whose language comprehension
was at the three-year level and whose
language expression was even more retarded
in spite of an overall IQ of 78 and of performance abilities at above-age level. Some
participants noted only the child's cognitive
difficulties in their diagnosis. However, even
those participants who made a primary
diagnosis of a specific disorder of language
for this boy did not agree in their coding,
some coding under 306 ("special symptoms
not elsewhere classified"), some under 308
("behavior disorders of childhood"), and
some under 310 ("mental retardation"). This
case also emphasized the lack of instructions
on how to deal with multiple diagnoses. It
also demonstrated the lack of a suitable coding for developmental disorders (also noted in
1967 at the Paris Seminar). This arose partly
through a lack of suitable categories in the
ICD and partly through the scattered and illdefined nature of such categories.
Considerations of the Paris Seminar Report
Concerning Child Psychiatry
In discussing the case histories it was noted
several times that the same problems were
being encountered as those previously discussed at the Paris Seminar on child psychiatric disorders. The Paris Seminar (4) had
noted that there was no adequate provision
for child psychiatric disorders in ICD-8, and
decided that any scheme for the inclusion of
such disorders must be simple and practical,
must include only a basic minimum of
information (in this respect classification
necessarily differed from both a diagnostic
formulation and nomenclature), and must
include unambiguous coding since the chief
purpose of a classification is to facilitate
communication.
Two main issues had arisen out of the case
histories presented at the Paris Seminar: 1)
Some psychiatric diagnoses for children were
not included in the current ICD, and 2) the
same diagnoses were coded differently by
different psychiatrists because of a lack of
explicit instructions about what the coding
should include. As a result of these issues the
Paris Seminar (4) recommended a triaxial
scheme of classification in which all three
axes had to be coded. The first axis was
"clinical psychiatric syndrome," the second

was "intellectual level," and the third was

"associated and etiological factors."
The present Seminar recommended that
the proposals of the Paris Seminar for the
classification of child psychiatric disorders be
accepted and noted that the classification of
mental retardation posed similar problems.
It was agreed that a similar scheme was
needed for mental retardation but that some
modifications might be required.
Principles of Classification
It was agreed that the diagnosis of mental
retardation necessarily involved recording
several different and independent aspects of a
case. Thus, at a minimum, it was essential
that the degree of retardation, as well as the
basic disorder (where present), be classified.
Obviously, it would be totally unsatisfactory
if it were only possible to record that a patient
had either Down's syndrome or moderate retardation. In fact, both statements are required for a diagnosis of his condition. In
addition, many mentally retarded patients
show some major emotional or behavioral
disorder that must be recorded for the purposes of providing medical care.
It was decided that this problem could be
solved by either a multiaxial scheme, as
proposed at the Paris Seminar, or by a multicategory scheme. In both cases a clear set of
instructions would have to be provided to
ensure that each axis or category was routinely recorded for all patients. It was decided
that at least three major axes or categories
were required, namely, those outlined in the
Paris Seminar. For all mentally retarded patients the intellectual level, the associated or
etiological factors, and the clinical psychiatric syndrome would need to be recorded.
Whether a multiaxis or a multicategory
scheme is adopted would be dependent on
decisions made by WHO in relation to the
organization of the ICD as a whole.
The three classes of information needed to
provide a satisfactory classification of mental
retardation correspond with those proposed
for child psychiatry, and the seminar foresaw
no great problem in working out a scheme
that would serve the needs of both disciplines
while retaining compatibility with the rest of
the ICD. The essential feature of the scheme
is that for each patient three categories of
information should be provided. Codings
should be available that note where informa-

tion is not known or where no abnormality is

present.
It was noted that the necessity for clear
instructions on how to code multiple diagnoses is one common to all parts of the ICD
that concern mental disorders. It was recommended that a similar multiaxial or
multicategory system be considered for the
psychiatric section of the ICD as a whole.
Degree of mental retardation. It was
agreed that the degree of mental retardation
should constitute a principal dimension in the
diagnostic classification. Discussion centered
around three issues in this connection: 1) the
criteria to be used in the assessment of retardation, 2) the level below which retardation would be regarded as present, and 3) the
subdivision into degrees of retardation.
It is now known that intelligence is not a
fixed and immutable quality, and in the present state of knowledge, prognostications
about future intellectual development are
necessarily rather uncertain. In view of these
considerations it was agreed that, in line with
the recommendations of the Paris Seminar,
mental retardation should be assessed on the
basis of current level of functioning without
regard to its nature or causation.
It was also agreed that retardation concerned intellectual functioning and that social
handicaps due to other disorders, e.g., sensory
defects or physical handicaps, should not be
included. When used appropriately, intelligence tests could provide valuable guidelines
to assess the level of retardation. However,
IQ scores should not be used in isolation;
rather, they should be taken in conjunction
with clinical judgments regarding the
patient's social and adaptive behavior and
development.
It was noted that intelligence tests were the
least useful and, indeed, might sometimes be
quite inappropriate for individuals from
populations in which the social and cultural
factors were quite different from those found
in the populations on which the tests were
standardized. Attention was drawn to the
Resolution on Psycho-Technical Tests ( 7 )
passed by the Educational, Scientific, Cultural, and Health Commission of the Organization of African Unity (OAU). This reads as
follows:
CONSIDERING that the psycho-technical
tests at present used in our countries are illadapted and do not correspond with our culture,

our environment and our development;

CONSIDERING the danger of giving wrong
guidance to our youth and workers through
continued use of these ill-adapted tests;
CONSIDERING the importance of having tests
that are adapted to studying and using our human
resources to better advantage; RECOMMENDS
1) that studies be undertaken by Member States,
wherever possible, to establish properly adapted
psycho-technical tests; 2) that OAU, with the assistance of the United Nations Specialized Agencies (UNESCO: WHO: FAO: UNICEF) and the
ICC take part in these studies.The Seminar considered the Fifteenth Report of the WHO Expert Committee on
Mental Health on the "Organization of
Services for the Mentally Retarded" (8). It
recommended that its suggestions on classification by degree of mental retardation be
accepted. The Committee had criticized the
classification of those with an IQ in the range
of 68-85 as being "borderline mentally retarded," noting that this vastly widened the
concept of mental retardation, in that at least
16 percent of the general population would be
considered retarded. The Committee also
expressed itself as being strongly opposed to
this expansion of the concept, taking the view
that a level of functioning equivalent to an IQ
two standard deviations below the mean, i.e.,
about 70, was a most useful upper demarcation of mental retardation. The Seminar
expressed the view, however, that IQ limits
should constitute only a guide, it always being
necessary to take clinical considerations of
social and adaptive functioning into account.
The Seminar concurred with the recommendation of the Expert Committee that the
term "borderline mental retardation" had no
place in a medical classification and that it
should be dropped from the ICD. As it was
necessary in all patients to make some coding
under the rubric of "current level of intellectual functioning," it was recommended that
the coding of "borderline mental retardation" be replaced by a coding of "normal
variations in intelligenceincluding borderline intelligence."
The Seminar concurred with the recommendations of the Expert Committee on the
various degrees of mental retardation, name-

ly that the categories of "mild," "moderate," ry of "associated and etiologic factors" use,
"severe," and "profound" be retained. How- where appropriate, ICD codings from other
ever, in view of the advice that IQ scores sections. The organization of this method of
should never be the sole measure of degree of coding needs further consideration and it may
retardation, it was recommended that IQ be necessary to provide special codings for
levels that define the categories be omitted definite neurological disorders that do not
from the category headings. Instead, the constitute a clearly defined disease of a recterms should be carefully defined in the ognized type. A working party needs to be
Manual on Psychiatric Disorders and Classi- set up to determine how this should best be
fications. Where appropriate, the IQ ranges done. Alternatives to be considered by the
proposed by the Expert Committee should be working party are that the fourth digits
used as a guide instead of the current limits should merely indicate the presence of an asICD-8 suggests.3 However, the IQ limits sociated physical condition that would then
should constitute just one aspect of the defini- be coded under its ICD number, or that the
tion of categories that should include a care- provision for coding such disorders under the
ful description of the degree of handicap in fourth digits be revised to provide a more
social and adaptive terms.
satisfactory system. The list of available
The category "unspecified mental retarda- terms in ICD-8 should be reviewed to ensure
tion" (315 in ICD-8) should be retained but that all diagnoses required for the satisfactoinstructions to coders should indicate very ry classification of mental retardation were
clearly that it be used as sparingly as pos- available and that added provision be made
sible; it is intended solely for patients whose where necessary. The Seminar recommended
current level of intellectual functioning can- that the terms describing conditions comnot be assessed either by standardized tests monly found and reported in the classificaor by clinical judgments (e.g., a newborn).
tion of mental retardation be brought toAssociated or etiological organic factors. gether in a glossary accompanying the
As the case history exercise showed, the classification.
fourth-digit coding for mental retardation
Occasionally more than one associated
proved to be quite unreliable. This is partly organic condition may be present. For exbecause it demands a knowledge of the etiol- ample, one of the patients in the exercise had
ogy of the retardation, which is often lacking diabetes, as well as epilepsy, but only a small
due to the pathogenesis of mental retardation number of the participants coded diabetes. It
being only imperfectly understood. Further- was recommended that, as a rule, the diagmore, nine of the fourth digits combine into nosis of the condition most closely associated
groups a larger number of conditions repre- with the pathogenesis of mental retardation
sentative of many areas of ICD-8, so that as a be recorded. Where feasible, and where the
statement of etiology, they are inadequate. patient's condition demands this for purposes
Moreover, clinicians or coders who work of medical care, more than one diagnosis
mainly with Section V of ICD-8 do not al- should be entered on the second axis. (The
ways have the complete manual available to problems of dealing with data involving
them; hence they may be unable to code cor- multiple coding present no difficulties in
rectly conditions other than those specifically modern computer technology, but coding
mentioned as inclusions of the fourth digits as more than one diagnosis may present problisted, and in some cases, they may fail to re- lems to the personnel involved in maintaining
cord relevant information on diagnosis.
records systems.) Where there are no organic
The Seminar recommended that these dif- features associated with the patient's reficulties be eliminated by making the catego- tardation, this fact should be recorded on the
second axis.
Associated or etiological psychosocial factors. Problems of intellectual retardation
arise not infrequently in relation to psychosocial factors, and the Seminar considered
it desirable that there be provisions for the
coding of such factors. The provision and
definition of categories of psychosocial

influences posed difficulties beyond the scope

of the present Seminar, but it was recommended that a working party4 be set up to
develop appropriate definitions for psychosocial factors, both those important in the
pathogenesis of mental retardation and also
those influences, familial and other than familial, that are important in the pathogenesis
of emotional and behavioral disorders. In
view of the importance of psychosocial
influences it was recommended that this be a
separate axis or category to be recorded for
all patients instead of the present fourth digit,
.8, that associates mental retardation with
psychosocial (environmental) deprivation and
that was felt to be insufficient.
Genetic factors in mental retardation.
Some cases of mental retardation are due to
specific diseases that are genetic in origin.
These should be noted in the category recording etiological or associated physical
conditions. In addition, however, cases of
mental retardation not due to any brain
disease often result from an interaction between polygenic factors and environmental
influences. Although the science of behavioral genetics is rapidly advancing, the Seminar recognized that in the current state of
knowledge it is usually not possible to differentiate genetic influences from psychosocial
influences of an environmental kind.
Associated psychiatric
and behavioral
conditions. The Seminar agreed that coding
any associated psychiatric condition constituted an essential part of the diagnosis and
classification of mental retardation. An axis
or category should be included to deal with
this dimension. As far as adult patients were
concerned, Section V of ICD-8 provided
suitable categories, and as far as child patients were concerned the recommendations
of the Paris Seminar should be accepted.
Classification of Child Psychiatric Disorder
It was noted that the recommendations on
child psychiatric disorders necessitated only a
few extra codings and that provision might be
made for these by transferring the ICD-8
category 308 and utilizing categories 316 to
319 that are at present not assigned. By appropriate adjustments to the glossary to take

account of disorders in children, neurotic

disorders, personality disorders, psychosomatic disorders, and other clinical
syndromes could be included in existing codings. By redefining psychoses and by providing extra digits, child psychoses could also be
included under the current codings. Normal
variation, conduct disorder, and manifestation of mental subnormality only would need
additional codings. Adaptation reaction
would need an extra coding, but this might be
provided by a redefinition and reorganization
of category 307, "transient situational disturbances." Specific developmental disorders
also need a special category but this might
be provided by a reorganization of category
306.
Mental Retardation in Adults
Whereas the Seminar spent the majority
of its time discussing mental retardation in
children, it was recognized that any classification scheme must also apply to adult patients. It was thought that the scheme suggested by the Seminar would be equally
appropriate for all age groups.
Glossary and Instructions on Use
Throughout its deliberations, the Seminar
stressed the need for a glossary that would
bring together terms commonly used in describing mentally retarded patients, whether
or not these were found in Section V of ICD-8
or in other sections. An essential task is to
define terms. Several countries and professional organizations have produced glossaries
concerning mental retardation, that put out
by the American Association on Mental
Deficiency being the most comprehensive (9).
The Seminar welcomed the initiative of
WHO, which has undertaken to produce a
glossary of mental disorders, taking into account the existing national glossaries. A publication incorporating a glossary should also
contain a manual that would give clear coding instructions about what should be
included in, and excluded from, any particular category of the classification. Where the
instructions state that a particular diagnosis
should not be included in a particular category, there should be clear instructions about
where the diagnosis should be coded.
Consistency is essential; this cannot be
achieved unless coding instructions are unam-

biguous and unless they cover most contingencies.

Classifications for Different Purposes
The Seminar agreed that no classification
would meet all purposes: Statistical data on
mental retardation are of interest not only to
clinicians, but also to geneticists, other medical scientists, and psychologists, educators,
and social service agencies concerned with
health, education, and welfare. The Seminar
noted that whereas the ICD was originally
designed to provide a basis for vital statistics
and for public health purposes, increasing attention was paid in the future to health service
needs, including the utilization of hospital
and other medical care facilities. However, it
was recognized that a classification that
served these needs would not necessarily be
entirely satisfactory to educators and to social agencies concerned with welfare, or in the
treatment of offenders who did not present
psychiatric or other medical problems. While
it took cognizance of these problems, the
group recognized that they fell outside the
scope of medicine; and for the purposes of the
ICD it is important that categories included
in it be relevant to medical needs. If classifications for different needs are produced, they
should be capable of translation into the ICD
categories. However, no classification should
be used unless it has been satisfactorily tested
in practice, and in general the use of different
classifications for different purposes should
be discouraged.
Recommendations to Other Working Groups
The Seminar noted that the current ICD-8
classification of neurological disorders associated with mental retardation is not entirely
satisfactory, for example, in relation to epilepsy and to certain types of encephalopathy
not diagnosed as specific diseases. The Seminar called the attention of the Working
Group on Neurological Disorders to this
shortcoming. Neurological disorders often
accompany mental retardation, and the Seminar expressed the wish that the working
group should be cognizant of the problems
of mental retardation in its deliberations.
Promoting the Effective Use of ICD
The Seminar discussed what steps could be
taken to ensure that the ICD be used in the
most effective manner possible, both within
a given country and internationally. It ex-

pressed its appreciation of the lead that

WHO had taken in this field and stressed the
vital part it could play in promoting further
studies.
The point was made that a decision by a
member country or by a professional organization to use the ICD routinely for reporting purposes could have a beneficial
effect not only upon the standard of case reporting, but also upon the attention paid to
diagnosis and classification in medicine. If,
for example, pediatricians were constantly
reminded of the need for early diagnosis and
classification, this would lead not only to
their making better use of the ICD, but it
would also influence the attention paid to
diagnosis and classification during medical
and pediatric training. Use of the ICD might
thus have an influence upon medical education in a more general sense.
Attention should also be paid to providing
medical students with training in classification and in the use of the ICD scheme. This
topic should be included in the curricula of
clinical training. Case history exercises might
be a suitable method of teaching in this connection.
At a local level there needs to be close collaboration among biostatisticians, clinicians,
and coding officers in medical records departments regarding the use of the ICD. Records officers require training in the use of the
ICD and of the manual of coding instructions, and they should be encouraged to work
closely with clinicians and to return to them
for clarification records that do not permit
ambiguous coding. Seminars and short courses in the correct use of the ICD according to
the manual and glossary would do much to
improve the quality of statistical reporting;
and regular feedback both for queries regarding particular patients and for material
fed to a central statistical office would ensure
that record keeping achieves and maintains
high standards.
Professional organizations (including local
and national medical societies and associations for the scientific study of mental retardation) can also help to educate their
members in the use of the ICD. Studies
should be planned and carried out with the
help of statisticians who should be consulted
early and with whom analyses of data should
be discussed at each stage of inquiry-research
and fact-finding. At an international level the

support of the International Association for

the Scientific Study of Mental Deficiency
should be sought.
The Seminar was impressed by the usefulness of the diagnostic case reports and videotape exercises in bringing to light specific
problems and in clarifying concepts.
National Centers of Health Statistics and
professional organizations should be encouraged to work together to adapt the ICD
to more specialized purposes, thus ensuring
that when they themselves are involved in
collecting statistical data, the data would be
put in a form that would allow the use of the
ICD.
Even in countries capable of carrying out
case reporting and videotape exercises on
their own, the initiative of WHO in sponsoring such exercises by bringing together
experts from different countries and in providing case materials and videotapes from
different countries had been of great importance in efforts to promote uniformity in
diagnosis and in case reporting. It was recommended that WHO should promote further exercises of this type at local and
regional levels. It was also hoped that WHO
would be able to make readily available
case materials from different centers in
different countries and that it would further
promote or facilitate seminars concerned
with diagnosis and classification.
Arising from the deliberations at hand, the
Seminar emphasized the importance of preparing at an early date and of testing in practice a provisional classification on the lines
suggested earlier. The results of any field
studies that use the proposed classification
and that should, by preference, be carried out
in more than one country, should be reported
back to WHO. A future seminar that would
take up the problems raised but not settled in
this one would be invaluable. The group recognized that WHO could take a lead in stimulating these developments. Indeed, without
WHO's sponsorship they are unlikely to occur at all.

Seminar on Psychiatric Diagnosis, Classification, and Statistics, that dealing with child
psychiatry (4). This would require that for
each patient, the following four types of
information would be recorded: 1) degree of
mental handicap, 2) etiological or associated
biological or organic factors, 3) associated
psychiatric disorder, and 4) psychosocial factors. For each patient, all four types of
information would be routinely reported,
instead of only the degree of mental handicap.
Degree of mental retardation. In the assessment of the degree of mental retardation,
relevant information about the sociocultural
background of a patient and his social and
adaptive functioning must be taken into account. The grade of mental retardation recommended by the Expert Committee on
Mental Health (8) should be used in the ninth
revision of ICD. These comprise ICD categories 311-314"mild," "moderate," "severe," and "profound" mental retardation,
together with category 315"unspecified
mental retardation." Category 310, "borderline mental retardation," which includes
backwardness, borderline intelligence, deficientia intelligentiae, borderline mental
deficiency, or subnormality, and an IQ range
of 68 to 85 should be replaced by a category
of normal variations in intelligence in ICD-9.
The Seminar departed from the recommendations of the WHO Expert Committee
on Mental Retardation in recommending
that IQ ranges should not be included in the
ICD manual, but rather, should be specified
in an accompanying glossary that would draw
attention to the limitations, as well as to the
usefulness, of IQ data for the assessment of
intellectual handicaps. The glossary should
also stress that in evaluating the grade of
intellectual retardation, social and cultural
background be taken into account.
Organic aspects. The second type of
information to be recorded for each patient
should, at a minimum, consider the principal
organic feature, if any, associated with the
retardation. If no such features are reported,
Summary and Recommendations
this should be recorded. Users should employ
multiple coding on this axis, etiological and
The Ninth Revision of ICD
The Seminar considered alternative ap- other diagnoses being included where approproaches to the problems of classification in priate.
Psychiatric and behavioral aspects. The
mental retardation. It decided in favor of a
scheme compatible with, and derived from, third class of information should include
the proposals recommended by the Third psychiatric symptoms or syndromes catego-

rized in a form compatible with that used in

child psychiatry and elsewhere in the ICD.
The same considerations regarding multiple
coding that apply to the previous section
should apply to this one.
Psychosocial factors. The Seminar recognized the importance of psychosocial factors
in the pathogenesis of mental retardation and
recommended that psychosocial influences be
recorded on a separate axis or category. A
working party is needed to develop appropriate categories and to provide definitions
for them.
Adequacy of ICD Rubrics
A small working group should review the
rubrics used in sections of the ICD other than
Section V to see whether they will provide on
the axes of the proposed classification a
comprehensive list of terms for classification
in mental retardation. This working group
should include representatives from the seminar on child psychiatry to ensure compatibility between the two systems of classification, and between them and the rest of the
ICD.
The same working group should discuss
with WHO the possibility of incorporating
sociocultural factors in the diagnostic classification, linking this with the attempt now
being made by WHO to classify socioenvironmental factors leading to hospitalization
or to a need for medical care.
Recommendations Concerning Other Scientific Groups
Because mental retardation is so frequently
associated with neurological disorders,
including epilepsy, the Seminar expressed the
hope that the scientific group considering
neurological disorders would bear in mind
the problems of mental retardation in making their recommendations. It was also hoped
that the forthcoming seminars concerned
with character disorders and with neuroses
would give some consideration to any particular problems arising in relation to mental
retardation.
Glossary and Manual of Instructions
The Seminar was greatly impressed by the
need for a glossary5 and for a manual of

instructions to accompany ICD-9. It welcomed both WHO initiative and the efforts of
national and of professional organizations in
taking steps to provide an acceptable glossary. The group recommended that the glossary should bring together terms commonly
applicable to the mentally retarded and found
in all sections of the ICD, and that the accompanying manual of instructions should
give guidance to users concerning where particular diagnoses should or should not be
placed.
Promoting the Effective Use of the ICD
The Seminar recommended that WHO
should consult with government agencies and
international and professional organizations
about steps that might be taken to promote
the effective use of the ICD in member
countries. These steps include: further diagnostic exercises, national and regional seminars organized on the lines successfully pioneered in the series of WHO seminars on
classification, and short training courses for
medical students and for persons particularly
responsible for coding and classification.
The Need for Field Trials
The Seminar recommended that WHO
should consider as soon as it can the possibility of sponsoring field trials in different
countries in which the proposed classification
would be tried in practice. The results of any
field studies should be reported back to
WHO. It was also recommended that a future meeting be convened to discuss the results of the field trials and also the integration
of the various recommendations of the seminars on different mental disorders.
The Seminar endorsed the recommendations of the Paris Seminar with regard to the
need for an adequate provision of categories
regarding child psychiatric disorders in ICD9.
The 1969 recommendation of the Educational, Scientific, Cultural, and Health
Commission of OAU (7) was noted with approval. The recommendation proposes that
appropriate intelligence tests be developed
for different cultures and that tests developed
for one culture should not be applied without
modification in very different cultures. The
Seminar considered that the same issue might
apply to different cultures within one
country.

REFERENCES
1. World Health Organization: Report of the Sixth
Seminar on Standardization of Psychiatric Diagnosis, Classification, and Statistics. Geneva, WHO,
1971
2. Shepherd M, Brooke EM, Cooper JE, et al: An
experimental approach to psychiatric diagnosis. Acta Psychiat Scand 44 (suppl 201), 1968
3. Astrup C, Odegard O: Continued experiments in
psychiatric diagnosis. Acta Psychiat Scand 46:180209, 1970
4. Rutter M, Lebovici S, Eisenberg L, et al: A tri-axial
classification of mental disorders in childhood. An
international study. J Child Psychol Psychiat
10:41-61, 1969
5. Averbuch ES, Melnik EM, Serebrjakova ZN, et al:
Diagnosis and Classification of Psychiatric Disease
in Old Age. Leningrad, Institute of Psychiatry, 1968
6. World Health Organization: International Classification of Diseases, 8th revision. Geneva, WHO, 1968
7. Educational, Scientific, Cultural, and Health
Commission, Organization of African Unity: Resolution on psycho-technical tests. Passed at the first
ordinary session, Addis Ababa, Ethiopia, June
30-July4, 1969
8. Organization of services for the mentally retarded.
WHO Techn Rep Ser 392, 1968
9. Heber R: A manual on terminology and classification in mental retardation, 2nd ed. Amer J Ment
Defic 65 (monograph suppl), April 1961
APPENDIX 1
List of P a r t i c i p a n t s
M e m b e r s of Nuclear G r o u p
Dr. Jack R. Ewalt, Bullard Professor of Psychiatry, Harvard Medical School, and Superintendent, Massachusetts Mental Health Center,
74 Fenwood Rd., Boston, Mass. 02115, U.S.A.
Dr. Masaaki Kato, Division of Adult Mental
Health, National Institute of Mental Health,
Konodai, Ichikawa City, Chiba-Ken, Japan
Dr. Morton Kramer, Chief, Biometry Branch,
National Institute of Mental Health, 5454 Wisconsin Ave., Chevy Chase, Md. 20014, U.S.A.
Dr. Ornuly O. Odegard, Medical Superintendent, Overlege Ved Gausted Sykebus, Vinderen,
Oslo, Norway
Dr. H. Rotondo, Professor of Psychiatry, San
Fernando Medical School, San Marcos University, Lima, Peru6
Dr. Michael L. Rutter, Institute of Psychiatry,
De Crespigny Park, Denmark Hill, London
S.E. 5, England
Dr. Raymond Sadoun, Directeur de Recherche,
Institut National de la Sante et de la Recherche
Medicale, 3 rue Leon-Bonnat, 75 Paris 16e,
France
Dr. Z.N. Serebrjakova, Chief Specialist in Psy-

choneurology, Ministry of Health of the U.S.S.R.,

Moscow, Rahmanovsky per 3, U.S.S.R.
Dr. M. Shepherd, Professor of Epidemiological
Psychiatry, Institute of Psychiatry, University of
London, De Crespigny Park, London S.E.5,
England
Dr. A.V. Sneznevskij, Director, Institute of
Psychiatry of the Academy of Medical Sciences of
the U.S.S.R., Moscow B-152, Zagorodnoe
Schosse 2, U.S.S.R.
Dr. Hans Strotzka, Professor of Medicine, 16/
24 Daringergasse, Vienna 1190, Austria (ViceChairman)
Dr. George Tarjan, Professor of Psychiatry,
School of Medicine and School of Public Health,
University of California at Los Angeles, and Program Director, Mental Retardation, the Neuropsychiatry Institute, 760 Westwood Plaza, Los
Angeles, Calif. 90024, U.S.A. (Chairman)
P a r t i c i p a n t s from t h e United States
Dr. Philip R. Dodge, Professor of Pediatrics and
Neurology and Chairman, Department of Pediatrics, Washington University, and Physician-inChief, St. Louis Children's Hospital, 500 S. Kings
Highway, St. Louis, Mo. 63110, U.S.A.
Dr. Leon Eisenberg, Professor of Psychiatry,
Harvard Medical School, and Psychiatrist-inChief, Massachusetts General Hospital, Fruit St.,
Boston, Mass. 02114, U.S.A. (Rapporteur)
Dr. Julius B. Richmond, Chairman and Professor of Pediatrics, and Dean, College of Medicine,
Upstate Medical Center, 766 Irving Ave., Syracuse, N.Y. 13210, U.S.A.
Dr. Joseph Wortis, Director of Developmental
Services and Studies, Maimonides Medical
Center, 4802 Tenth Ave., Brooklyn, N.Y. 11219,
U.S.A.
P a r t i c i p a n t s from O t h e r Countries
Dr. Augusto Aguilera, Director, Mental Health
Department, Ministry of Public Health, and Associate Professor of Psychiatry, University of San
Carlos, 5 Ave. 8-33, Zona 4, Guatemala City,
Guatemala, Central America
Dr. Stanislau Krynski, Professor of Child Psychiatry, and Director, Mental Retardation Institute, Federal University, Escola Paulista Medicina, Sao Paulo, Brazil
Dr. Denis Lazure, Professeur agrege de psychiatrie, University de Montreal, P.O. Box 6128,
Montreal 3, Quebec, Canada
Dr. Hernan Montenegro, Child Psychiatrist,
Department of Pediatrics, Hospital Roberto del
Rio, Universidad de Chile, Martin Alonso Pinzon
6702, Santiago de Chile, Chile (Rapporteur)
Dr. Carlos E. Sluzki, Director, Centro de
Investigaciones Psiquiatricas, Paraguay 1373,
Buenos Aires, Argentina

Dr. Dario Urdapilleta, Director, Children's

Psychiatric Hospital, Ingenieros 32-101, Mexico
City 18 D.F., Mexico
T e m p o r a r y Advisors
Dr. Tsung-yi Lin, Professor of Mental Health
and Psychiatry, University of Michigan, Ann Arbor, Mich. 48106, U.S.A.
Dr. Jack Tizard, Professor of Child Development, Institute of Education, University of
London, Malet St., London W.C.I, England
Secretariat
Dr. Michael J. Begab, Head, Mental Retardation Research Centers Program, National Institute of Child Health and Human Development,
Bethesda, Md. 20014, U.S.A.
Ms. Eileen Brooke, Mental Health Consultant,
World Health Organization, 1211 Geneva 27,
Switzerland
Dr. Felix de la Cruz, Special Assistant for
Pediatrics, Mental Retardation Program, National Institute of Child Health and Human
Development, Bethesda, Md. 20014, U.S.A.
Dr. Rene Gonzalez, Regional Advisor in
Mental Health, Pan American Health Organization, 525 Twenty-third St., N.W., Washington,
D.C. 20037, U.S.A.
Dr. B.A. Lebedev, Chief, Mental Health Unit,
World Health Organization, 1211 Geneva 27,
Switzerland

Dr. W.P.D. Logan, Director, Division of Health

Statistics, World Health Organization, 1211
Geneva 27, Switzerland
Dr. Norman Sartorius, Medical Officer, Mental
Health Unit, World Health Organization, 1211
Geneva 27, Switzerland (Secretary)
Official Observers
Dr. Sterling Garrard, Professor of Pediatrics,
Upstate Medical Center, State University of New
York, 750 E. Adams St., Syracuse, N.Y. 13210,
U.S.A.
Dr. Robert E. Luckey, Consultant, Program
Services, National Association for Retarded
Children, 420 Lexington Ave., New York, N.Y.
10017, U.S.A.
Dr. I.M. Moriyama, Health Services and
Mental Health Administration, National Center
for Health Statistics, Washington, D.C. 20201,
U.S.A.
Mr. David B. R a y , E x e c u t i v e D i r e c t o r ,
President's Committee on Mental Retardation,
Department of Health, Education, and Welfare,
Rm. 3760, Washington, D.C. 20201, U.S.A.
Ms. Winthrop Rockefeller, President's Committee on Mental Retardation, Department of
Health, Education, and Welfare, Rm. 3760,
Washington, D.C. 20201, U.S.A.
Dr. Paul T. Wilson, Principal Investigator,
Information Processing Project, American Psychiatric Association, 1700 18th St., N.W., Washington, D.C. 20009, U.S.A.

Some Thoughts on the Classification of

Mental Retardation in the United States of America
BY GEORGE TARJAN, M.D., AND LEON EISENBERG, M.D.

It is desirable to reach a generally accepted

international resolution that assures that patients afflicted with severe emotional disorders and manifesting the symptomatology of
mental retardation are classified into the
same category. The solution advocated by the
current Diagnostic and Statistical Manual of
Mental Disordersplacing first emphasis
on mental retardationoffers the most for
comparability
of biostatistical
information
from diversified geographic settings.
Further,
the
manual's
encouragement
of multiple
psychiatric diagnoses assures against loss of
information.

HE MOST COMMONLY USED classification

system in mental retardation in the

United States of America is the one contained in the Manual on Terminology and

Classification in Mental Retardation, second

edition, of the American Association on
Mental Deficiency (AAMD)(1). Its categorization of medical diagnoses is similar to
that of the Diagnostic and Statistical Manual
of Mental Disorders, second edition {DSMII), published by the American Psychiatric
Association (APA) (2). Both manuals define
mental retardation in a similar fashion.
According to AAMD, "Mental retardation
refers to subaverage general intellectual
functioning which originates during the developmental period and is associated with
impairment in adaptive behavior" (1, p. 3).
APA's definition substitutes for "impairment
in adaptive behavior" the following phrase:
"impairment of either learning and social
adjustment or maturation, or both" (2,
p. 14).
We will first briefly discuss the parameters
of classification of mental retardation in
common use in the United States of America; then we will focus on two issues of diagnosis that are often subjects of controversy in
the United States.
P a r a m e t e r s of Classification of Mental
Retardation
The diagnosis of mental retardation in the
United States usually takes into account two
dimensions: medical classification based on
assignment to one of a number of specified
syndromes and severity of retardation based
on standardized developmental or intelligence tests. Both of these dimensions are
present in the AAMD and APA manuals. In
these respects the two manuals are very similar, except for the fact that AAMD places
first emphasis on medical classification and
DSM-II emphasizes degree of retardation.
Both medical classifications assign a primary
role to causation in their definitions of
syndromes and use syndromes of a descriptive nature only when etiology is unknown.
The AAMD classification system uses one
major dimension not contained in DSM-II,
i.e., classification by degree of impairment in
adaptive behavior, based on global clinical
judgment, in comparison to descriptive vignettes of age-specific adaptive behavior
deficits. Systematized ratings can also be
made in more than 20 supplementary categories, among which are genetic factors, impairments of special senses, psychiatric symp-

toms, motor dysfunctions, cultural conformity, and reading and arithmetic skills. The
utilization of all dimensions in the AAMD
classification system provides, in addition
to basic typology, a reasonably adequate
profile of the patient.
Controversial Issues of Diagnosis
Sociocultural Retardation
The first of the two special issues we wish
to discuss pertains to sociocultural retardation, a category frequently used in the United
States but not in some other countries. In the
AAMD classification two diagnostic categories are applicable to this group within the
major class of "mental retardation due to
uncertain (or presumed psychologic) cause
with the functional reaction alone manifest."
They are: "cultural-familial mental retardation" (code no. 81) and "psychogenic mental
retardation associated with environmental
deprivation" (code no. 82). In DSM-II the
appropriate category is mental retardation
"with psycho-social (environmental) deprivation" (subdivision .8), which includes two
subclasses: "cultural-familial mental retardation" and mental retardation "associated with environmental deprivation." Two
decades ago the diagnoses usually assigned to
the same group of patients were "familial"
and, less frequently, "undifferentiated" mental deficiency (3).
A general description of the patients who
qualify for these diagnoses follows. In most
instances retardation is of a mild degree, with
IQs in the range of 50 to 70. The condition is
usually not diagnosed prior to the individual's
entrance into school, and the overt diagnosis
generally disappears when he reaches
adulthood. Thus most patients are of school
age, i.e., six to 18 years old. An important
cause of the age specificity of the diagnosis
results from the basic clinical definition of
mental retardation in the United States,
which requires that subaverage general intellectual functioning and impairment in adaptive behavior be present concurrently. The
correlation between these two impairments is
highest during school years, when academic
demands make evident deficits that may not
be apparent when the practical skills of the
patient function adequately in the job market.
These generally mildly retarded patients

are normal in appearance and show no concomitant physical disabilities or abnormal

laboratory findings. The morbidity and mortality rates of the group are fairly average.
Children with economically, socially, and
educationally underprivileged backgrounds
have a high risk for this diagnosis. A conservative estimate places the risk of mild
retardation, including sociocultural retardation, at a level 15 times higher for impoverished urban and rural children than for those
of middle-class suburban origin (4).
Debates are still common about the etiology of sociocultural retardation. Children so
retarded are usually born to mothers who
were undernourished during their adolescence
and whose pregnancies occurred at a young
age and with high frequency. The prematurity
rate is twice the national average. Prenatal
care is either nonexistent or limited, and
perinatal and postnatal care is below average.
During infancy and early childhood the patients' nutrition is often inadequate. The
children are exposed to a series of somatic
noxae, including infections, poisons, and
traumata, and they are often unprotected by
customary public health measures. It is
therefore not difficult to conceptualize a
variety of biomedical models (5) that explain
the causation of this type of retardation on
the basis of the cumulative effects of organic
insults to the central nervous system. Some
clinicians prefer such biomedical models,
while others proffer a genetic explanation for
this syndrome on the theory of assortative
mating of the mildly retarded.
But these are also children who are usually
unwanted, unplanned, and conceived accidentally and, frequently, extramaritally.
They are raised in homes with absent fathers
and with physically or emotionally unavailable mothers. During infancy they are not
exposed to the same quality and quantity of
tactile and kinesthetic stimulations as other
children. Often they are left unattended in a
crib or on the floor of the dwelling. Although
there are noises, odors, and colors in their
environment, the stimuli are not as organized
as those found in middle-class and upperclass environments. For example, the number
of words they hear is limited, with sentences
brief and most commands carrying a negative
connotation. From these empirical observations a causal model of environmental deprivation has been constructed by American

behavioral scientists, and the syndrome acquired its labels: sociocultural deprivation,
psychosocial deprivation, cultural-familial
retardation, etc. (6).
Two current etiological questions that have
a bearing on diagnostic classification need
discussion. The first involves the role of organic factors in the causation of sociocultural
retardation. The AAMD manual describes
mental retardation due to "uncertain (or
presumed psychologic) cause with the functional reaction alone manifest" as suited only
for those instances of mental retardation that
occur "in absence of any clinical or historical
indication of organic disease or pathology
which could reasonably account for the retarded intellectual functioning" (1, p. 39).
"Cultural-familial mental retardation" and
mental retardation "associated with environmental deprivation," into which groups
patients with sociocultural retardation are
placed, are two subclasses of this major
category. If clinicians were to adhere literally
to the category description, they might not
diagnose anyone as socioculturally retarded
because, in most instances, the history of poor
medical care and the multiple exposures to
somatic noxae in themselves would contradict the requirement of absence of historical
indications of organic disease.
Observations on the longitudinal development of impoverished children and on the
effects of major changes in the mode of rearing favor the psychosocial etiological model.
On the other hand, the uncritical acceptance
of pure functional causation does not take
into account the probable effects of the biomedical traumata. As a consequence, the erroneous conclusion might be drawn that
somatic noxae do not play a role in the
causation of sociocultural retardation.
The second etiologic question involves the
specific roles of the various elements of deprivation. Global conclusions have been
drawn concerning the total effect of deprivation without specific information on the components of deprivation in regard to quality,
quantity, specificity, or timing.
Transcultural studies, in addition to solving diagnostic and classification problems,
have much to offer toward a better understanding of the causation of sociocultural retardation (7). More specifically, much is to be
gained from such studies about the roles of
genetic, somatic, and experiential forces as

they interact in producing that mild, and

age-specific, type of retardation that has a
high prevalence in impoverished population
groups and that is generally labeled in the
United States as sociocultural retardation.
Relationship Between the Diagnoses of Early
Childhood Psychoses and Mental Retardation

tism and who, on the basis of evaluations of

his intellectual and adaptive performances,
qualifies for a diagnosis of retardation in this
framework, would be placed into the category
of "pseudo-retardation."
The controversy about the interrelationship
between psychosis and retardation, when both
coexist, is not yet resolved. The AAMD classification provides two categories for these
patients: code no. 83, "psychogenic mental
retardation associated with emotional disturbance," and code no. 84, "mental retardation associated with psychotic (or major
personality) disorder." The underlying assumption is that mental retardation in these
patients is due to psychologic causes, and
therefore the diagnosis of mental retardation should be made with the presumed psychologic causation being specified. DSM-II
provides one major category for this group:
mental retardation "following major psychiatric disorder" (subdivision .7). The
underlying philosophy is essentially the same
as that found in the AAMD manual.
It is unquestionable that severe mental
disturbances in infancy and early childhood
impair intellectual adaptive performances.
As a consequence, such patients fulfill the
requirements of the diagnosis of mental
retardation by both AAMD and APA standards. Moreover, measurable IQ in such patients functions as the best single prognosticator of outcome (9). On the other hand, one
might argue that the primary disease in these
children is the psychosis or the emotional
disorder, with mental retardation being one
of the several manifestations of the severe
emotional pathology. At the present time no
firm scientific conclusions can be drawn on
the basis of etiologic research. The argument,
although often heated, therefore remains a
philosophic and semantic one.

The second controversial problem pertains

to the relationship between the diagnoses of
early childhood psychoses and mental retardation. It is not uncommon to find children in the United States who sequentially,
and in any combination, acquire a series of
diagnoses that include early infantile autism,
mental retardation, childhood schizophrenia,
brain damage, early childhood autism, and
minimal brain dysfunction. At times the
clinical pictures in these patients are further
complicated by a variety of organic or functional sensory impairments.
On the surface this problem may appear
quite limited in scope, but closer scrutiny
calls attention to a number of ramifications.
One example of the consequences of diagnostic and classificatory uncertainty is the
varied use of the term "pseudo-retardation"
among clinicians in the United States (8).
Some use the term to describe the socioculturally retarded, i.e., those in whom no central nervous system damage can be demonstrated; others restrict the term to those
in whom sensory or motor deficits produce
some of the symptoms of retardation. Advocates of these concepts often argue that in
these cases true retardation (or deficiency),
per se, is not present and that the impairment
in measured intelligence is more a function of
the inadequacies of the psychological tests
than of the impairment in the patient's cognitive functions.
Some clinicians prefer to use the term
"pseudo-retardation" in those instances in S u m m a r y
which, in their judgment, retarded intellecFor the benefit of national and internatual performance and inadequate adaptive
behavior are explainable on the basis of tional professional communications, it is
underlying
psychogenic
mechanisms. desirable to reach a generally accepted res"Pseudo-retardation" in this context refers to olution that assures that patients who are afthose patients in whom mental retardation flicted with severe emotional disorders and
results from psychosis, severe emotional dis- who manifest the symptomatology of mental
turbance, neurotic disorder, or other types of retardation are classified into the same catemajor personality disorders during infancy gory, independent of the idiosyncrasies of
and early childhood. For example, the child clinicians. The solution advocated by DSMwho during infancy manifests the signs of au- II has the most to offer to bring order into

this semantic chaos. It states: "'Mental retardation is placed first to emphasize that it is
to be diagnosed whenever present, even if due
to some other disorder" (2, p. 1). The manual's encouragement of multiple psychiatric
diagnoses assures against loss of information.
The decision of the manual might seem arbitrary, but it offers the most for comparability
of biostatistical information from diversified
geographic settings.

4.

5.
6.

7.

REFERENCES
1. Heber R: A manual on terminology and classification in mental retardation, 2nd ed. Amer J Ment
Defic 65 (monograph suppl), April 1961
2. American Psychiatric Association: Diagnostic and
Statistical Manual of Mental Disorders, 2nd ed.
Washington, DC, APA, 1968
3. American Association on Mental Deficiency: Sta-

8.
9.

tistical Manual for the Use of Institutions for Mental Defectives, 3rd ed. New York, National Committee for Mental Hygiene, 1946
Tarjan G: Some thoughts on socio-cultural retardation, in Social-Cultural Aspects of Mental Retardation. Edited by Haywood CH. New York, AppletonCentury-Crofts, 1970, pp 745-758
Tarjan G: The next decade: expectations from the
biological sciences. JAMA 191:226-229, 1965
Eisenberg L: Social class and individual development, in Crosscurrents in Psychiatry and Psychoanalysis. Edited by Gibson RW. Philadelphia, JB
Lippincott Co, 1967, pp 65-88
Cravioto J, Delicarde ER, Birch HG: Nutrition,
growth, and neurointegrative development. Pediatrics 38:319-372, 1966
Eisenberg L: Emotional determinants of mental
deficiency. Arch Neurol Psychiat 80:114-121, 1958
Rutter M: The influence of organic and emotional
factors on the origins, nature and outcome of childhood psychosis. Develop Med Child Neurol 7:518528. 1965

Differing Concepts of Diagnosis

as a Problem in Classification
BY JACK R. EWALT, M.D.

The author discusses the differences in the

British and U.S. concepts of mental retardation that make achieving a uniform international classification difficult. Although U.S.
and British definitions and classifications
of mental retardation seem to be similar, a
basic conflict exists. The British define mental
retardation as an arrested or incomplete
development of the brain, while the United
States defines it as a person's mental status
current at a given time but that may be
subject to change.

AJOR PROBLEMS in reaching agreement

on an international classification of
diseases may be attributed to: 1) differences
in the perception of symptoms, and 2) differences in the inference and interpretation of
the meaning of a symptom in diagnostic
terms. These problems have been amply il-

lustrated by international diagnostic exercises (1) and by a joint exercise of the United
States and the United Kingdom (2).
International uniformity in classification
may be easier to achieve for the disorders
known as mental retardation or mental subnormality than for the psychoses, neuroses,
and character disorders. However, differences
in the concepts and interpretations of disordered behavior between countries must be
c o n s i d e r e d in our a t t e m p t s to r e a c h
agreement on a scheme of classification for
mental retardation. In the British Glossary of
Mental Disorders (3), mental retardation is
divided by degree of severity into six categories; these categories appear to be the same as
those in the eighth revision of the International Classification of Diseases (ICD-8) (4)
and as those in the second edition of the Diagnostic and Statistical Manual of Mental
Disorders (DSM-II) (5). All of these systems
define their subdivisions by estimates of the
levels of retardation expressed in terms of
numerical IQs, as well as in descriptive
phrases ranging from " b o r d e r l i n e " to
"severe." This apparent agreement between
British and U.S. classifications is not firm,
however, since it seems that the manner in
which the level of retardation is assessed and

the attitude about prognosis vary; herein lies

the problem in comparative statistics.
British Concepts
The British define subnormality (by quoting from the England and Wales Mental
Health Act of 1959) as "a state of arrested or
incomplete development of mindwhich
includes subnormality of intelligence and is
of a nature or degree which requires or is
susceptible to medical treatment or other
special care and training of the patient" (3,
p. 22). Severe subnormality is defined as "a
state of arrested or incomplete development
of mind which includes subnormality of
intelligence and is of such a nature or degree
that the patient is incapable of living an independent life or of guarding himself against
serious exploitation or will be so incapable
when of age to do so" (3, p. 22). They also
state that the term "severe subnormality"
does not always relate exactly to that position in their six-point classification. They
discuss the difficulties of assessing IQ levels
and how test scores will vary among different
IQ tests.
U.S. Concepts
DSM-II uses similar terms in defining
mental retardation, but it is apparent that
these terms do not mean exactly the same
thing as the British ones. DSM-II states that:
"Mental retardation refers to subnormal
general intellectual functioning which originates during the developmental period and
is associated with impairment of either learning and social adjustment or maturation, or
both" (5, p. 14). It also states:
It is recognized that the intelligence quotient
should not be the only criterion used in making a
diagnosis of mental retardation or in evaluating its
severity. It should serve only to help in making a
clinical judgment of the patient's adaptive behavioral capacity. This judgment should also be based
on an evaluation of the patient's developmental
history and present functioning, including
academic and vocational achievement, motor
skills, and social and emotional maturity (5, p.
14).
In the United States, the most widely used
' glossary for mental retardation is A Manual
on Terminology and Classification in Mental Retardation. 2nd edition (6), of the Amer-

ican Association on Mental Deficiency

(AAMD). The DSM-II section dealing with
mental retardation is essentially a modification of the manual in order to make it conformto ICD-8.
The AAMD classification has two large
categories, which are organized in DSM-II as
subgroups (fourth-digit items) under each of
the five degrees of retardation. The two categories are 1) Medical Classification, and
2) Behavioral Classification.
Medical Classification includes the following groups: 1) diseases due to infection;
2) diseases due to intoxication; 3) disease due
to trauma or physical agents; 4) diseases due
to disorder of metabolism, growth, or nutrition; 5) diseases due to new growths; 6)
disease due to unknown prenatal influence;
7) disease due to unknown or uncertain
causes, with structural reactions manifest;
and 8) disease due to an uncertain cause with
the functional reaction alone manifest and
presumed psychologic.
Under Behavioral Classification, behavior
is considered in two dimensionsmeasured
intelligence and adaptive behavior. Thus, according to the manual:
Mental retardation refers to subaverage general
intellectual functioning which originates during
the developmental period and is associated with
impairment in adaptive behavior.... The definition specifies that the subaverage intellectual
functioning must be reflected by impairment in
adaptive behavior. Adaptive behavior refers primarily to the effectiveness of the individual in
adapting to the natural and social demands of his
environment. Impaired adaptive behavior may be
reflected in: (1) maturation, (2) learning, and/or (3)
social adjustments. These three aspects of adaptation are of different importance as qualifying
conditions of mental retardation for different age
groups (6, p. 3).
The manual also mentions that objective
measures for adaptive behavior, especially for
degree of maturation and for quality of social
adjustment, are not sufficiently developed to
allow easy assessment.
Currently, one can only make estimates of
impairment in adaptive behavior by comparing the patient's performance with that of
persons of the same age level in the general
population. The objective measures of general
intelligence must be ". .. supplemented by
evaluation of the early history of self-help and
social behavior, by clinical evaluation of pres-

ent behavior, and by whatever measures of

academic achievement, motor skills, social
maturity, vocational level, and community
participation are available and appropriate"
(6, p. 4). Impairment in any one of these
spheres reflecting the subaverage intellectual
function is all that is required for a diagnosis
of mental retardation. However, it is rare that
a person shows mental retardation in only
one of these categories.
At first glance, these statements are in essential agreement with those in the British
glossary. However, the A A M D glossary
continues:
Within the framework of the present definition,
mental retardation is a term descriptive of the
current status of the individual with respect to
intellectual functioning and adaptive behavior.
Consequently, an individual may meet the criteria
of mental retardation at one time and not at
another. A person may change status as a result of
changes in social standards or conditions or as a
result of changes in efficiency of intellectual functioning, with level of efficiency always being determined in relation to the behavioral standards
and norms for the individual's chronological age
group (6, p. 4).
This latter attitude is somewhat in conflict
with the British definition: "a state of arrested
or incomplete development of mind which
includes subnormality of intelligence and is
of such a nature or degree that the patient is
incapable of living an independent life or of
guarding himself against serious exploitation
or will be so incapable when of age to do
so" (3).
Summary
The difference in these concepts of mental
retardation will cause little if any confusion
in the classification of patients with severe
structural changes of the brain. Confusion

will arise over the less severely retarded and

in the classification of persons who have been
successfully rehabilitated and who are functioning satisfactorily in an ordinary environment. The U.K. use of the term, which, to
the best of my knowledge, somewhat accurately reflects the general attitude on the
Continent, describes prognostic qualities of
the condition and in some way reflects their
attitude toward the term "psychoses." In the
United States, on the other hand, the term
"mental retardation" describes a combination of behaviors manifest in an individual at
the time of examination, and while the term
has some prognostic implications, a bad
prognosis is not seen as an essential element
in arriving at the diagnosis. This attitude may
cause an overinclusion of patients in the U.S.
series, especially patients in whom retardation is due to educational, language, and social problems, and whose retardation might
be largely corrected by proper environmental
and educational experiences, but who at the
time of assessment were definitely retarded in
their mental and and social functioning.
REFERENCES
1. Shepherd M, Brooke EM, Cooper JE, et al: An
experimental approach to psychiatric diagnosis. Acta Psychiat Scand 44 (suppl 201), 1968
2. Kramer M: Cross-national study of diagnosis
of the mental disorders: origin of the problem. Amer
J Psychiat 125 (April suppl): 1-11, 1969
3. Subcommittee on Classification of Mental Disorders
of the Registrar General's Advisory Committee on
Medical Nomenclature and Statistics: A Glossary of
Mental Disorders. London, Her Majesty's Stationery Office, 1968.
4. World Health Organization: International Classification of Diseases, 8th revision. Geneva, WHO, 1968
5. American Psychiatric Association: Diagnostic and
Statistical Manual of Mental Disorders, 2nd ed.
Washington, DC, APA, 1968
6. Heber R: A manual on terminology and classification in mental retardation, 2nd ed. Amer J Ment
Defic 65 (monograph suppl), April 1961

Comments on the ICD Classification of

Mental Retardation
BY JOSEPH WORTIS, M.D.

Strictly speaking, mental retardation is not a

disease but a symptom; it may be the result of
biological deficits, vicissitudes of experience,
or both. The author believes that ways must
be found to avoid the misleading implications
and consequences of dealing with it as a
disease. Toward this end, he proposes a multidimensional system of classification that
distinguishes between the biologic and psychosocial causes of mental retardation.
ISTORICALLY the International Classification of Diseases {ICD) began as a
classification of the causes of death; it was
later extended to include a classification of
diseases. Mental retardation is not a cause of
death and can scarcely be called a disease.
The historical origin of the classification of
mental retardation was of a different nature
and related mainly to the needs of educators
to group their pupils on the basis of attained
levels of mental development. These origins
are connected with the names of Binet and
Simon, both of whom dissociated themselves
from the tendency (most pronounced in the
United States) to equate IQ scores with innate biological capacity. Half a century ago
they wrote:

As a general rule, the children classed as retarded are the victims of disease, constitutional debility
or malnutrition. We find included in our lists some

who are the children of migratory parents; some

who have been kept from school; some who have
attended a religious school, where they learned
little but sewing and writing; some who have
changed their school too often; some also who are
foreigners and understand little French, and lastly,
some who have been kept back in their studies by
unrecognized myopia ... (1, p. 47).
T h e Problem with Standardized Tests
Since one's level of intellectual performance, or intelligence, is always a product of
both biological equipment and educational
experience, and since social and educational
opportunity cannot be said to be uniformly
available to all, it is unreasonable to assume
that intelligence is normally distributed on a
Gaussian curve. Yet standardized tests, like
the Stanford-Binet Intelligence Scale, have
been trimmed and altered in an attempt to
achieve just such a distribution, although
some s k e w n e s s to the left c a n n o t be
avoided (2). A test constructed in such a way
will inevitably disclose that a minimum of 15
percent of the population utilized (i.e., IQs
two standard deviations from the mean) is
mentally retarded. But since IQs vary with
social class, the percentage will be much
higher in the lower classes. The American
Negro population, most of whom are poor
and socially deprived, was excluded from the
Stanford-Binet standardization; it is not surprising therefore that at least half of the
American Negro child population, whose
mean IQ is currently about 85, could be stigmatized with the disease or condition called
mental retardation, by the criteria (one standard deviation) of the eighth revision of ICD
(ICD-8)(3). In the Puerto Rican rural population the percentage is even higher (4). These
absurdities are compounded when a test
whose standardization is based on a white,
somewhat middle-class population is applied
to the population of a foreign nation.
The other horn of the dilemma is this: If

the test is restandardized and is thus adapted

to a foreign nation, under the assumption that
the intelligence of that nation's populace is
normally distributed, one will again find that
about 15 percent of the population is mentally retarded, since a standardized test must
yield roughly that percentage one or more
standard deviations from the mean. Comparative statistics thus become meaningless
and even historical changes within the same
country elude analysis, since the tests must be
constantly restandardized.
An IQ score, at best, can indicate where an
individual stands in intellectual performance
compared to others. What others? His nation? His social class? His ethnic group? No
intelligence test that has ever been devised
can surmount all of these complicating
considerations and claim universal validity.
International criteria based on IQ scores thus
have no validity, and are, at best, rough indicators of one's relative rank in roughly comparable groups, for example, in the urban
populations of New York City and of
London. One may question whether any important scientific interest is served in ICD-8
by the use of IQ criteria for the diagnosis of
mental retardation; and one may ask whether
these interests could not be served more
modestly, yet more effectively, by the use of
well-defined clinical judgments as indicators
of general adaptive capacity.
Shortcomings of C u r r e n t Classifications
The principle that different purposes require different classifications is clearly recognized in the introduction to ICD-8 (3). A
classification designed for educational needs
should be based on different criteria and
categories than a medical classification is
based on. Is there not a certain confusion of
aims involved in our present classification?
ICD-8 lists five categories under "mental
retardation" (310-314) that are based on
IQs, spaced by standard deviations from the
mean, on intelligence tests developed under
the assumption that intelligence is normally
distributed. There is also a category for "unspecified mental retardation" (315). There
are nine subcategories (.0-.9) based on classes
of etiology, i.e., infections, trauma, metabolic
disorders, brain disease, prenatal influences,
chromosomal abnormalities, prematurity,
major psychiatric disorder, psychosocial

causes, and other and unspecified. These

subcategories are not complete and are not
mutually exclusive, logically consistent, or
systematic; in some instances mental retardation "follows" the presumptive cause, in
others it is "associated with" the cause, and in
others it is "due to" the cause. This pragmatically accords with the general pattern of
ICD-8, which makes no claim to completeness, logical consistency, mutually exclusive categories, or systematic organization, although it has attempted to move
toward these goals. Current usage must be
reckoned with, and a practical classification
has to be a compromise between usage and
ideals.
An important source of some of the difficulties in classification derives from the
failure to make distinctions among etiology,
true pathology, physiological malfunction,
level of performance, and capacity for
development (5).
Virchow's theory of cellular pathology
served a useful function in its day but has
been partly superseded by the concept of
pathophysiological dysfunction as the basis
for disease. In other words, it is not the anatomical lesion or defect that causes the disease
or disability but the resulting physiological
malfunction. Some anatomic defects may not
significantly impair function, and conversely,
some malfunctions may be due to chemical,
electrophysiological, or other deficiencies not
associated with anatomical disease. Sometimes the anatomical lesion is only a late
product of the malfunction. To take an extreme example, an exhausted, overstimulated, poorly nourished child may have serious and chronic learning difficulties because
of his condition, but, strictly speaking, he is
not diseased and his condition is reversible,
up to a point. Beyond this point irreversible
pathophysiological malfunction may supervene and at a later stage pathological anatomical lesions may appear.
Suggestions for a New Classification
Since mental retardation is not a disease
but a level of development or a level of function, perhaps it would be better to list it as a
symptom in a subcategory rather than as a
primary disease diagnosis. From this point of
view, the suggestion of a triaxial system of
diagnosis in child psychiatry has merit (6).

Perhaps a lesson can be learned from the

American Heart Association (7), which decided long ago to employ the following fourdimensional system of cardiac diagnosis: 1)
etiological (e.g., rheumatic); 2) anatomical
(e.g., mitral stenosis); 3) physiological (e.g.,
auricular fibrillation); 4) functional (e.g.,
dyspneic at rest); and in addition to these: 5)
possible heart disease; and 6) potential heart
disease.
If such a system were applied to mental
retardation, performance level could be
measured by an intelligence test or an adaptive scale, while physiological efficiency could
be measured by Pavlovian paradigms, evoked
potentials, electroencephalograms, expectancy curves, intersensory transfer, inhibitory
and discriminatory capacity, or other
measures of basic cognitive function that are
relatively uncontaminated by vicissitudes of
experience. The situation would be quite different if we had some measure of cerebral efficiency at levels low enough to permit objectivity, but high enough to be relevant to
intellectual processes. While we do not have
such reliable measures, perhaps it is not too
early to begin to use all available data to
make judgments on the physiological intactness or efficiency of an individual's cerebral
functions as a basis for appropriate management. Such an approach would help sharpen
our diagnostic skills and should serve to
break down the harmful assumptions of
homogeneity implied in our current dependence on IQ scores.
Intelligence tested to show an IQ above 50
correlates closely with social class. Mild and
borderline retardation is largely an accompaniment of poverty (8, 9). It seems highly
desirable to make diagnostic formulations
that would distinguish more sharply between
biologic mental retardation and psychosocial
mental retardation. By using the same primary diagnosis for both types of mental retardation, as is the current trend, this distinction tends to be obscured, and dissimilar
problems with dissimilar needs are likely to
be lumped together, as is the case in our
schools.
Because there is a practical difference between the two broad categories of biological
handicap and of psychosocial deprivation, I
think the earlier method of distinguishing
between the two groups should be restored,
even when well-validated diagnoses cannot be

made (10). If this were done, there would be a

category for known or presumed biological
defect and a category for presumed biological
intactness. Appropriate terms or numerals
could be applied to each. Thus a postencephalitic patient with an IQ of 65 would
no longer carry the same primary diagnosis
as that of a biologically normal slum dweller
with the same IQ. In many cases a presumptive diagnosis of biological handicap would
have to be made on the basis of a compromising medical history, delayed developmental
milestones, motor awkwardness, soft
neurological signs, and empirical teaching
experience with the child, all leading to an
assumption of, biological inadequacy with no
further specification possible. As our diagnostic skill improved this nonspecific category would in time be reduced as more definitive diagnoses became possible.
From an empirical point of view, the most
common labels I apply to patients are: 1)
encephalopathy, cause unknown; 2) encephalopathy, presumed or definite cause;
3) mental retardation, cause unknown, presumed biological; 4) mental retardation,
cause unknown, presumed psychosocial; and
5) mental retardation, cause unknown, presumed mixed causes.
Of the specific disease entities that are
diagnosable, mongolism is by far the most
common and is likely to be found in about 15
to 20 percent of the patients seen in a medical
setting with a presenting complaint of retardation (11). Encephalopathy is likely to be
diagnosed in about 45 percent but etiology
can seldom be established, and in about 15
percent it is not even possible to assume the
presence of an encephalopathy. About 15
percent of these patients have other primary
psychiatric disorders. All other specific medical causes make up, at most, a small percentage of the retarded population, using
currently available diagnostic knowledge and
resources. But in spite of their rarity, specific
diagnoses should be encouraged, and the frequency of such diagnoses should increase. In
former years much emphasis was placed on a
presumed idiopathic form of mental retardation as an aspect of normal variation in native
intelligence. With our increased awareness of
biological causation on the one hand and
psychosocial causation on the other, the frequency of this diagnosis has declined and in
some settings it is rarely made. Where it oc-

curs it can either be subsumed under the

category "mental retardation, cause unknown, presumed biological," or, less desirable, a category of idiopathic mental retardation can be created. On the whole we do
not get very far with our current classification, since so many patients are not specifically diagnosable. With the exception of
mongolism, most diagnoses are still based on
clinical judgments that have strong subjective
influences.
Conclusion

ICD-8 realistically and pragmatically disclaims logical consistency, completeness, and

even modernity. It attempts to note all current designations for diseases or conditions
requiring medical attention and to supply
labels and numbers for them in order that
they can be counted. It thus uses eponyms,
anatomical lesions, metabolic disorders, and
a number of overlapping categories so that a
given condition or disease can often be labeled in several different ways. From this
point of view the acceptance by ICD-8 of the
condition called mental retardation is a
simple reflection of contemporary practice.
In ICD-8, Section XVI is devoted to Symptoms and Ill-Defined Conditions and lists
such things as coma, sleep disturbances,
speech impediments, fainting, hiccoughs,
nausea, sweating, fatigue, unknown fever,
malingering, and the like. There is even a
category for "mental observation." From this
point of view mental retardation, although
not a disease, could also be included here. It
would be better, however, if it could be more
clearly designated as a symptom rather than
as a disease. The primary diagnosis would
then be the disease, if diagnosable, or a category of diseases, such as encephalopathy, if it
could not be further diasnosed. I would like

to recommend the creation of four broad

major categories that would help in our approach to these cases: 1) presumed biologic
cause, 2) presumed psychosocial cause, 3)
presumed mixed cause, and 4) unknown
cause. Where a concurrent presence of a true
disease and of mental retardation exists,
some suitable designation should be devised
to indicate whether or not there is a presumed
causal relationship between the two.
REFERENCES
1. Binet A, Simon T: Mentally Defective Children.
Translated by Drummond WB. New York, Longman, Green Co, 1914, p47
2. Terman LM, Merrill MA: Stanford-Binet Intelligence Scale, 3rd ed. Boston, Houghton-Mifflin Co,
1960
3. World Health Organization: International Classification of Diseases, 8th revision. Geneva, WHO, 1968
4. Albizu-Miranda C, Matlin N: Psychosocial aspects
of cultural deprivation, in Proceedings of the First
Congress of the International Association for the
Study of Mental Deficiency. Edited by Richards
BW. Reigate, England, M Jackson Publishing Co,
1968, pp 466-474
5. Heber R: A manual on terminology and classification in mental retardation, 2nd ed. Amer J Ment
Defic 65 (monograph suppl), April 1961
6. Rutter M, Lebovici S, Eisenberg L, et al: A tri-axial
classification of mental disorders in childhood. An
international study. J Child Psychol Psychiat
10:41-61, 1969
7. American Heart Association: Diseases of the Heart
and Blood Vessels: Nomenclature and Criteria for
Diagnoses, 6th ed. Boston, Little, Brown and Co,
1964
8. Ginzberg E, Bray DW: The Uneducated. New York,
Columbia University Press, 1953
9. Gruenberg EM: Epidemiology, in Mental Retardation. Edited by Stevens HA, Heber R. Chicago,
University of Chicago Press, 1964, pp 259-306
10. Mental Retardation: Improving the Classification
Through Research. Work conference sponsored by
the American Association on Mental Deficiency,
Columbus, Ohio, Sept 23-25, 1965 (processed)
11. Wortis J, Wortis H: Who comes to a retardation
clinic? Amer J Public Health 58:1746-1752, 1968

A Note on the International Statistical Classification

of Mental Retardation
BY JACK TIZARD, PH.D.

The author feels that classification in mental

retardation should be multiaxial. While it is
unimportant whether diseases or conditions
associated with it are included in Section V
(Mental Disorders) of the World Health Organization's International Classification of
Diseases or elsewhere, it is important that
terms be used consistently and that a glossary
that is comprehensive and operational in
definition accompany the classification manual.

whether of diseases or
of any other phenomena, will be used
only if it is found useful. To be really useful, a
classification of mental retardation must
serve a variety of purposes: medical (clinical,
genetic, and epidemiological) as well as social (giving guidelines for education and
training and for the planning of services).
A useful classification of mental retardation must accommodate children and adults.
It cannot be too elaborate or no one will use
it; yet it must also be both comprehensive and
compatible with other sections of the eighth
revision of the International Classification of
Diseases (ICD-8) (1).
CLASSIFICATION,

The Eighth Revision of the ICD

Although for some purposes a single axis
of classification is adequate, this is not true in
psychiatry. In mental retardation a single
axis is unsatisfactory because the mentally
retarded commonly have multiple disabili-

ties, none of which is primary, and because

the classification has to serve more than one
function. ICD-8 is better than previous revisions because its categorization is more logical and because its use of a fourth digit permits a biaxial system of classification. The
primary axis is a division by grade of defect;
five grades are distinguished (borderline, IQ
of 68 to 85; mild, IQ of 52 to 67; moderate,
IQ of 36 to 51; severe, IQ of 20 to 35; and
profound, IQ of less than 20).
This classification was criticized by the
WHO Expert Committee on Mental Health
(2) on two main counts. The first was that although IQ ranges were used to describe the
five categories, no indications were given of
the mean and standard deviation on which
these ranges were based. The second criticism
was that the classification of individuals with
an IQ in the range 68 to 85 as "borderline
mentally retarded" would vastly widen the
concept of mental retardation; on this basis,
16 percent of the general population would be
considered mentally retarded. The WHO
committee was strongly opposed to this
expansion of the concept of mental retardation; it held the view that an IQ of 70 (two
standard deviations below the mean) was the
traditional and most useful upper borderline
measure.
The WHO committee advocated slightly
different IQ ranges to define the various
grades of intellectual retardation: mild, 2 to
-3.3 standard deviations from the mean of
100, i.e., IQ of 50 to 70; moderate, 3.3 to
-4.3 standard deviations from the mean, i.e.,
IQ of 35 to 50; severe, 4.3 to 5.3 standard
deviations from the mean, i.e., IQ of 20
to 35; and profound, more than 5.3 standard deviations from the mean, i.e., IQ
of less than 20. It should be stressed, the
committee said, that these are not exact
measurements, and they should not be
considered the sole criteria for diagnosis.

In practice the categories will tend to overlap, but the IQ has some value within the
range of mental retardation both as a diagnostic and a prognostic guide.
Although some of the terms used in both
the ICD-8 (1) and the WHO report (2) to describe the gravity of intellectual handicaps
are perhaps somewhat sanguine, the actual
division by grade makes good clinical sense.
Unfortunately, however, the intention that all
mentally retarded persons should be categorized first by the severity of their intellectual
handicap may be nullified if ICD-8 category
315, "unspecified mental retardation," allows
clinicians to avoid making judgments about
the grade of defect. It would be preferable to
require that particulars of grade of defect be
recorded for all patients classified as mentally retarded, a clinical estimate being given
if no psychometric data were available.
It is therefore recommended that in future
revisions of ICD-8, category 315 should be
omitted. Every person classified as mentally
retarded should be assigned to one or other of
the grades finally agreed upon. (In the case of
infants the grading must be provisional, but
this could be specified in the instructions).
The second axis in ICD-8 (the fourth digit)
is broadly medical, and disorders are divided
into ten categories. Most of these are
reasonably satisfactory, although they are too
broad. (Category .2, for example, lumps together "disorders of metabolism, growth, or
nutrition," and category .5 is for those "with
chromosomal abnormalities.") It is recommended that the second axis be expanded to
two digits to allow for finer differentiation
within each category.
While several categories in the present
fourth-digit subdivision would probably
benefit from the results of discussion among
pediatricians and psychiatrists (e.g., category
.6, mental retardation "associated with prematurity"), there are two categories in particular that are important for behavioral
scientists. These are category .7 ("following
major psychiatric disorder") and category .8
("with psycho-social [environmental] deprivation").
Tarjan and Eisenberg discuss these problems in relation to the classification of mental
retardation in the United States of America
(3). They point out:
It is not uncommon to find children in the

United States who sequentially, and in any combination, acquire a series of diagnoses that include
early infantile autism, mental retardation, childhood schizophrenia, brain damage, early childhood autism, and minimal brain dysfunction. At
times the clinical pictures in these patients are
further complicated by a variety of organic or
functional sensory impairments. . . .
At the present time no firm scientific conclusions can be drawn on the basis of etiologic research [as to which is primary]. The argument, although often heated, therefore remains a philosophic and semantic one (3, p. 17).
Nonetheless, they feel that it is desirable
that children who have both severe emotional
disorders and symptoms of mental retardation be classified in a consistent fashion,
independent of the idiosyncrasies of clinicians. They themselves favor the solution recently advocated by the American Psychiatric Association (APA): "Mental retardation is placed first to emphasize that is to be
diagnosed whenever present, even if due to
some other disorder" (4, p. 1). This decision,
they say, might seem arbitrary, but it offers
the most for comparability of biostatistical
information from diversified geographic settings.
ICD-8 uses one of the fourth digits for this
type of patient (category .7, "following major
psychiatric disorder"). The word "following," however, prejudges the issue; the
term "with major psychiatric disorder"
would be better.
The other type of mental retardation that
Tarjan and Eisenberg discuss (3) is that
usually referred to as sociocultural retardation or, as in ICD-8 category .8, retardation
"with psycho-social (environmental) deprivation." Both the APA and the American
Association on Mental Deficiency (AAMD)
classifications (4, 5) divide sociocultural retardation into two categories: "cultural-familial mental retardation" and either mental
retardation "associated with environmental
deprivation" (APA) or "psychogenic mental retardation" (AAMD).
As Tarjan and Eisenberg point out, both
categories discount the possible effects of
biomedical traumata in bringing about functional retardation. "As a consequence, the
erroneous conclusion might be drawn that
somatic noxae do not play a role in the causation of sociocultural retardation. . . .
[Moreover] global conclusions have been

drawn concerning the total effect of deprivation without specific information on the
components of deprivation in regard to
quality, quantity, specificity, or timing" (3,
p. 16).
ICD-8's categorization of mental retardation "with psycho-social (environmental)
deprivation" is both clumsy and inaccurate.
"Psycho-social (environmental)," like "sociocultural" and "cultural-familial," tells us
no more than the older terms "familial,"
"undifferentiated," "residual," "aclinical,"
"subcultural," or "primary." The term "deprivation" is also a misnomer because it almost inherently contains the value judgment
that anyone who has not had a middle-class
upbringing is somehow deprived.
At the present time there is no way of
disentangling the effects of genetic, biological, and social factors in the causation of sociocultural retardation, and it would seem
more honest to acknowledge that fact
particularly since doing so is likely to lead to
a more reliable, and hence more useful, system of classification. Moreover, the degree of
environmental deprivation is usually judged
by the social circumstances of the family. A
more useful indicator of sociocultural retardation can probably be obtained from twoway tables that record the social class of the
parents and the clinical condition of the child.
Is "psycho-social (environmental) deprivation" worth recording as a cause of mental
defect at the present time?
Compatibility of the Various Sections
of ICD-8
The working group in child psychiatry of
the Third WHO Seminar on Standardization
of Psychiatric Diagnosis held in Paris in 1967
(6) opted for a triaxial system in which the
first axis described the "clinical psychiatric
syndrome," the second the "intellectual level," and the third the "associated or etiological factors." The group made little reference
to the classification of mental retardation in
ICD-8 and assigned mental retardation only
a single number9.0on the clinical axis. It
would be extremely unsatisfactory if diagnosis in child psychiatry were to remain in this
form, for the classification of a patient would
become a function of the place of referral or
the allegiance of a clinician.
This is, of course, the case today. ICD-8's
instructions say: "For primary mortality

classification and for morbidity classification

where the main interest is not in the mental
state, these categories [describing conditions
which are secondary to physical conditions]
should not be used, but assignment made to
the underlying cause." Does this mean that a
grossly defective spastic or epileptic child,
who is referred to a department of neurology
or pediatrics, is likely to be classified differently from the way in which he would be if
referred to a department of child psychology
or to a mental retardation clinic?
The problem of differential classification
according to place of referral may not have
been of great importance in the past because
of the lack of interest by departments of
pediatrics, neurology, and psychiatry in patients (especially children) with chronic
neurological and behavioral handicaps. But
this situation is changing, and it is important
to anticipate future developments. The problem is one that affects child psychiatry in
particular. Most mentally retarded patients
first come to notice as children, and for
planning purposes it is necessary to monitor
changes in prevalence and to highlight
inadequate case finding.
Hence an effort should be made to integrate the proposed triaxial system of classification proposed by the working group in
child psychiatry with that proposed for mental retardation. This could be done by
expanding the second axis of the triaxial
classification in child psychiatry (that concerned with intelligence, which, incidentally,
makes no separate provision for very bright
children) and by making the other two axes in
the classification proposed for child psychiatry compatible with those to be proposed for
mental retardation. The clinical axis in the
child psychiatry classification is more adequate than the corresponding axis in the
mental retardation section of ICD-8. The
etiological axes have much in common and
could probably be integrated, especially if a
two-digit system is to be usedas is proposed
in the child psychiatry working paper (6).
I hope that the seminar will consider how
classification in mental retardation can be
integrated with that in child psychiatry. A
multiaxial system has much to commend it
on theoretical grounds, but it would clearly
need to be tried out in practice before being
incorporated into the ICD. I therefore recommend that WHO assume responsibility

for pilot studies to provide information about

the feasibility of such a system.
C o m p r e h e n s i v e n e s s of C l a s s i f i c a t i o n
Whether or not a triaxial system common
to child psychiatry and mental retardation is
developed, consideration should be given to
the possibility of including additional physical handicaps from other sections of ICD-8 in
the mental retardation scheme. It is absurd,
for example, that cerebral palsy and epilepsy
should not find a place in a classification of
mental retardation.
For mental retardation, therefore, a useful
and feasible classification would require four
axes: 1) grade of intellectual functioning; 2)
etiological and medical diagnosis; 3) psychiatric aspects; and 4) additional physical
handicaps (to include epilepsy, cerebral
palsy, cleft palate, sensory handicaps, etc.).
It is easy to envisage that a system of classification could be devised that would be
common to child psychiatry, developmental
neurology, and mental retardation and that
would also serve the needs of the adult retarded. The inclusion of "associated physical
handicaps" would involve the introduction of
categories and terms that are rightly included
in other sections of ICD-8. As Wing (7) has
pointed out, this could be accomplished if the
classification were accompanied by a glossary, the need for which is manifest in any
case because of the confusion that surrounds
terminology in mental retardation. Wing
states:
It would therefore be useful for the Glossary to
include all sub-headings given in Section V of the
LCD. manual [so that it could be used by doctors
or clerical workers assigning codes without having
to search for appropriate labels in two different
books]....
A consistent policy for cross-references would
be desirable. . . . [and] . . . it would be most
helpful if the Glossary also included parts of the
LCD. manual, other than Section V, which might
be useful in classifying psychiatric problems (7).
In summary, classification in mental retardation should be multiaxial. While it is
unimportant whether diseases or conditions
associated with it are included in Section V of
LCD or elsewhere, it is important that terms
be used consistently and that a glossary that
is comprehensive and operational in definition accompany the manual.

REFERENCES
1. World Health Organization: International Classification of Diseases, 8th revision. Geneva, WHO, 1968
2. Organization of services for the mentally retarded.
WHO Techn Rep Ser 392, 1968
3. Tarjan G, Eisenberg L: Some thoughts on the classification of mental retardation in the United States
of America. Amer J Psychiat 128 (May suppl): 1418, 1972
4. American Psychiatric Association: Diagnostic and
Statistical Manual of Mental Disorders, 2nd ed.
Washington, DC, APA, 1968
5. Heber R: A manual on terminology and classification in mental retardation, 2nd ed. Amer J Ment
Defic 65 (monograph suppl), April 1961
6. Rutter M, Lebovici S, Eisenberg L, et al: A triaxial classsification of mental disorders in childhood. An international study. J Child Psychol Psychiat 10:41-61, 1969
7. Wing L: Observations on the psychiatric section of
the International Classification of Diseases and the
British Glossary of Mental Disorders. Psycho Med
1:79-85, 1970
Addendum
The first draft of this paper was sent for comment to a number of colleagues ' in Britain,
Europe, and the United States; many of their suggestions have been incorporated into the text.
Several colleagues felt that two aspects of the
classification system required more discussion.
Dr. Michael Begab, of the National Institute of
Child Health and Human Development, explained
this point of view as follows:
The terms "sociocultural," "culturalfamilial," and "psychosocial deprivation" are
indeed vague categories because of the uncertain
etiological factors underlying. Biological and
genetic contributors may well be involved but
their significance to intellect and function (when
too minimal to measure with current techniques) is yet to be established. In the absence of
more definitive substitutes, I doubt whether we
can do away with the only classification denoting the role of social-environmental circumstances as a cause of retardation. I am concerned this would lead to a conceptualization of
retardation as one deriving from biological determinants alone, thus drawing attention away
from the major dimensions of the problem. Although admittedly imprecise, the emphasis on

cultural deprivation has been an important

stimulant to educational, social and nutritional
intervention and research programs. If this
category were deleted in favor of an interactionist conceptminor neurological impairment,
genetic endowment and psychosocial deprivationwe may be back to an undifferentiated,
waste-basket classification.
My other point of contention is your implicit
discard of the supplementary adaptive behavior
classification. There is considerable disenchantment in this country with the IQ as an

index of current performance or future adaptation and we continue to struggle with more
refined measures of behavior. I doubt whether
the clinical classification proposed by the
Working Group on Child Psychiatry would
adequately cover this dimension. While I agree
that compatibility with other sections of the
ICD is important, the sacrifice to mental retardation is, in my view, too great. If we delete
this element because of limitations in
measurement, we may discourage research on
adaptive behavior scales.

The Problem of the Classification of

Mental Retardation
BY G. E. SUHAREVA, M.D.

The author proposes a new system of classifying the various forms of mental retardation. Using the time of exposure to a pathogenic agent and its etiology as a basis, she
classifies the forms of mental retardation into
three groups: 1) those caused by a pathological condition of the reproductive cells of the
parents; 2) those caused by harmful factors
that act during the intrauterine period; and 3)
those caused by damage to the central nervous system in the perinatal period or in the
first three years of life.

s A RESULT of the great progress that

has been made in the biological sciences
(luring the last few decades, a more solid
theoretical basis has been established for
studying the etiology and pathogenesis of
various forms of mental retardation.' Armed
with new and improved methods of research,
workers in various branches of theoretical
and clinical medicine have begun to study the
causes and mechanisms of anomalies in the
development of the human organism. During

this process wide use has been made of the

latest findings in medical genetics, embryology, biochemistry, and teratology and of
the clinical observations of obstetricians,
pediatricians, psychiatrists, psychologists,
and neuropathologists. The work of these
specialists has made it possible to establish
the etiology, pathogenesis, and clinical manifestations of many forms of mental retardation and to outline new ways of treating them.
In this way the latest findings on the etiology and pathogenesis of various forms of
mental retardation have enabled psychiatrists
to come closer to solving the intricate task,
set over 50 years ago, of breaking down this
complex group into its component clinical
forms.
Early Classifications
Attempts to differentiate among various
forms of mental deficiency were made by
psychiatrists as long ago as the 19th century.
Various criteria were proposed (for example,
for etiology whether a defect was congenital
or acquired and for the degree of intellectual
deficiency whether idiocy, imbecility, or feeblemindedness was present). However, at that
time there was no generally accepted classification, since the selection of a criterion depended largely on the purpose the classification was to serve. Thus, to determine the
possibility of teaching a mentally retarded

child or adolescent, use was commonly made

of a classification in which three degrees of
mental underdevelopment were distinguished. If feeblemindedness was present, the
child was sent to auxiliary schools, while patients who were imbeciles were taught to perform elementary forms of work. Persons at
the level of idiocy needed supervision and
care and could not adapt to an independent
way of life.
However, this method of settling the problem of mental retardation is not adequate for
properly organizing teaching of and therapy
with the mentally retarded or for determining
their working capacity and recommending
the best forms of work for them to do. Even if
their degree of intellectual deficiency is exactly the same, the working capacity of patients may differ depending on their level of
activity and drive. If there is marked asthenia
or apathy and if the intellectual defect is
combined with psychopathic manifestations,
the patient's activity is always reduced to a
greater or lesser extent. In the case of atypical and complicated forms of mental deficiency, the clinical picture of signs of local
defects, e.g., whether in hearing, in vision, or
in speech, and disturbances in cognitive activity and working capacity may be considerable, even though the patient's intellectual deficiency is slight.
For these reasons, coupled with the needs
of those who teach the mentally retarded and
the need for expert assessment of the working
capacity of the mentally retarded, use has
been made of a system of classification that
reflects the structure of the deficiency. An
outline of this kind has been proposed by
Pevzner(l). She distinguishes five clinical
forms of mental retardation:
1. An uncomplicated form without gross
deficiencies in any particular analyzer and
without marked emotional disturbances or
disorders of volition. This form is usually
hereditary in nature.
2. A form of mental retardation complicated by hydrocephalus in which the intellectual defect is combined with a behavioral
disturbance and a reduction in working capacity that includes an increased tendency to
tire quickly and attacks of headache. This
form is due to external causes.
3. Mental retardation combined with focal
disorders of hearing, speech, or the spatial
synthesis of the motor system.

4. Mental retardation with gross underdevelopment of the prefrontal areas of the brain
and that is characterized by specific changes
in personality and motor activity. This form
is of an exogenous etiology.
5. Mental retardation combined with
damage to the subcortical structures. This
form includes underdevelopment of the cognitive faculty and psychopathic behavior.
Need for a New Classification System
However, a different classification scheme
is required for the purposes of clinical practice and for scientific research; preferably,
this would be a classification system based on
the criterion of etiology and pathogenesis. A
criterion of this type for classifying mental
retardation has been used by many eminent
scientists who have studied the condition. Although these classifications reflect very well
the multiplicity of forms of mental retardation encountered in clinical practice, no
generally accepted classification exists. The
lack of unanimity on this question is not difficult to explain if it is remembered that the
very concept of mental retardation is interpreted differently by different workers.
For this reason I thought it necessary, before describing my classification scheme, to
insert a short introduction describing my
point of departure in defining the concept of
mental retardation and in delimiting it from
other clinical manifestations of intellectual
defect. In studying the clinical features of
mental retardation in children and in adolescents, I thought it essential to make a strict
distinction between the following two concepts: 1) an intellectual defect that is the
manifestation of the anomalous development
of the brain, and 2) intellectual disturbances
caused by damage to brain structures that
have already been formed. This distinction
fully accords with the ideas prevalent in teratology; the investigator sets himself the task
of distinguishing a developmental defect from
a disablement caused by damage to an organ
that has already been formed.
The clear-cut differentiation between these
two concepts makes it possible to distinguish
two forms of deficiency that differ in structure. The first, oligophrenic dementia, is a
nonprogressive, pathological condition that
constitutes a form of mental underdevelopment. The second is dementia in the sense of a

decay of mental functions that have already

been established and as a result of a process
that has produced the dementia.
I consider mental retardation to be a group
of pathological conditions with different
etiologies, but with one factor in common
all of the conditions represent clinical
manifestations of dysontogenesis, an anomaly of the development of the brain often
combined with developmental defects in
other body systems. Assigned to the group are
those forms of general mental underdevelopment characterized by: 1) the presence of a
defect in cognitive activity, and 2) the
nonprogressive nature of the condition.
It is important to emphasize that mental
retardation is characterized by the particular
complexity of its clinical manifestations; it is
a matter of the underdevelopment of the
highest forms of cognitive activity, which
cannot develop without the participation of
the ontogenetically and phylogenetically
youngest brain structures. These structures
mature late and are formed most intensively
during the first few years of postnatal development. That is why a disease process that
attacks the central nervous system of a child
during his first years of life can lead not only
to the destruction of previously formed systems, but also to the underdevelopment of
those structures that, at the time, have not yet
taken final shape. In view of this the forms of
mental retardation should include, in addition to the hereditary and congenital forms,
those forms acquired in the first few years of
life (up to three years of age).
If this concept of the essence of mental retardation were adopted, the boundaries of the
condition would become more clear-cut. The
forms should exclude, first, all those intellectual disturbances that occur at later stages in
the development of the child during various
progressive pathological processes affecting
the brain, or in the residual period, and that
represent the decay of intellectual functions
that have already been formed; and second,
milder forms of disturbances in intellectual
activity that are due to a slow rate of development (infantilism), incorrect child rearing,
asthenia from somatic causes, and behavior
disorders.
A T h r e e - G r o u p Classification
These were my initial assumptions when I
undertook a classification of mental retarda-

tion (appendix 1). Considering it to be a

special form of dysontogenesis of the brain,
and sometimes of the body as a whole, I
thought it essential to take into account the
laws governing the occurrence of developmental defects in general. Experimental research has shown that a developmental defect depends not only on the nature, intensity,
and acuteness of the pathogenic factor, but
also, and mainly, on the time of exposure,
i.e., the stage of ontogenesis at which the
organism was damaged.
For this reason two criteriathe time of
exposure and the nature of the pathogenic
agent (its etiology)were put forward as a
basis for differentiating mental retardation
into different clinical forms. In accordance
with this, all of the clinical forms of mental
retardation are divided into three groups,
depending on the time of exposure to the
harmful factor.
The first group is caused by a pathological
condition of the reproductive cells of the
parents, i.e., hereditary disease, a chromosomal aberration, and a pathological condition caused by exposure to harmful external
agents (ionizing radiation).
The second group is dependent on harmful
factors acting during the intrauterine period
(embryopathies and pathologies of the fetus).
The third group includes those forms of
mental retardation caused by damage to the
central nervous system in the perinatal period
or in the first three years of life, i.e., during
the period where the ontogenetically young
structures of the brain have not yet been
completely formed. Within each of these
three groups different clinical forms are distinguished on the basis of etiology.
Conclusions
The scheme proposed here for the classification of mental retardation cannot be
considered to be perfect and exhaustive. In
addition to the forms listed here, the causes of
which are more or less clear, there are a
number of other forms (the so-called undifferentiated forms of mental deficiency) for
which no accurate findings are available on
the causes and origins. The difficulty of differentiation on the basis of pathogenesis is
also due to the fact that some clinically welldefined forms of mental retardation have
been insufficiently studied with respect to

etiology. In addition, forms are quite often

seen in clinical practice that have multiple
causes, and it is difficult in each concrete case
to isolate the principal cause of the disease.
At the present time it is still not clear what
forms of enzymopathy can be considered to
be mental deficiences. It is often difficult to
make a differential diagnosis between an enz y m o p a t h y form of mental retardation and
dementia caused by a progressive enzymop a t h y disease. It can only be said that the
earlier the hereditary chemical defect is
discovered, the more often are observed
symptoms of the underdevelopment of cognitive activity of a mental deficiency type.
In other words, it is still difficult to determine a classification of mental retardation
that can be accepted as completely satisfactory. The only thing that is clear is the way we
should proceed if we wish to solve this problem in the future. The systematics of mental
retardation, as of other forms of disease,
must mainly be based on data regarding pathogenesis.
T h e p a t h o g e n e s i s of v a r i o u s t y p e s of
mental retardation depends not only on the
severity and nature of the etiological factor,
but also, and mainly, on the stage of ontogenesis at which the organism was damaged.
The more carefully we study the type of
reactivity of the nervous system at various
periods in antenatal and postnatal development, the easier it will be to establish a
classification of mental retardation, and the
better that classification will be.
REFERENCE
1. Pevzner MS: Mentally Retarded Children. Moscow,
Academy of Pedagogical Sciences Press, 1959
APPENDIX 1
T h e Classification of M e n t a l R e t a r d a t i o n
i n t o Three Groups
Group 1
Pathological Condition of t h e Reproductive
Cells of t h e P a r e n t s
Genetic Forms of Mental Retardation
1. Familial forms with a polygenic type of
inheritance
2. True microcephalus
3. Arachnodactylia (Marfan's syndrome)
4. Craniofacial dysostosis (Crouzon's disease)
5. Craniofacial dysostosis with syndactylia
(Apert's disease)

6. Laurence-Moon-Biedl syndrome
7. Mental retardation combined with the disturbance of endochondral ossification, with congenital epiphyseal dysplasia
8. Mental retardation combined with ichthyosis
(Rud's syndrome)
9. Some of the nevoid defects with a nonprogressive course
10. Mental retardation caused by damage to the
reproductive cells of the parents through
exposure to exogenous factors, e.g., ionizing
radiation
11. Other genetic forms
Enzymopathic F o r m s of Mental R e t a r d a t i o n
Disturbances of Protein Metabolism
1. Phenylketonuria (blockage of phenylalaninehydroxylase)
2. Maple syrup urine disease (disorders in the
metabolism of valine, isoleucine, and leucine)
3. Hyperlysinemia (disturbed metabolism of lysine)
4. Hypervalinemia (disturbed metabolism of
valine)
5. Histidinemia (disturbed metabolism of histidine)
6. Citrullinuria (disturbed metabolism of citrulline)
7. Homocystinuria (disturbed metabolism of
methionine)
8. Arginosuccinicaciduria (disturbed metabolism
of arginine)
Disturbances of Carbohydrate Metabolism
9. Galactosemia (a disturbance in the action of
the enzyme galactose-L-phosphate-uridyltransferase)
10. Fructosuria (hyperaminoaciduria)
11. Sucrosuria (intolerance of saccharose)
Disturbances in Pigment Metabolism
12. Methemoglobinemia (blockage of the enzyme
needed to convert methemoglobin into hemoglobin)
13. Deficiency of glucuronyl transferase and incapability of converting indirectly acting
bilirubin into the directly acting form (CriglerNajjar syndrome)
Clinical Forms of Mental R e t a r d a t i o n
Caused by C h r o m o s o m a l Aberrations
1. Mental retardation caused by a chromosomal
aberration in Group A chromosomes (ring
chromosomes)
2. Mental retardation caused by an aberration in
Group B that is connected with the deletion of
the short arm of the fourth pair of chromosomes (Wolfs syndrome)
3. Mental retardation connected with deletion of
the short arm of the fifth pair of chromosomes
("Cri du chat" syndrome)

4. Mental retardation connected with a trisomy

in Group D, 13th to 15th pairs of chromosomes (Patau's syndrome)
5. Mental retardation connected with an aberration in Group E; a trisomy of the 18th pair of
chromosomes (Edward's syndrome)
6. Mental retardation connected with deletion of
the short arm of the 18th pair of chromosomes
(De Grouchy's syndrome)
7. Mental retardation connected with deletion of
the long arm of the 18th pair of chromosomes
(Lejeune's syndrome)
8. Mental retardation caused by a trisomy of the
21st pair of chromosomes (Down's syndrome)
9. Mental retardation connected with an aberration in the system of sex chromosomes
(Klinefelter's syndrome)
10. Turner's syndrome
11. The triple X syndrome
12. Mental retardation in men connected with an
extra Y chromosome.
Group 2
The second group is comprised of types of
mental retardation caused by harmful factors acting during the intrauterine period. Making a distinction between the various clinical forms of
mental retardation on an etiological basis is
considerably more difficult in this group than in
the previous one, since it is not always possible to
determine which pathogenic factor is preventing
the establishment of the optimum environment for
the development of the embryo and the fetus (the
supply of nutritive substances and of oxygen).
These pathogenic factors may be different at different stages in intrauterine development. There is
no doubt that disturbances in uteroplacental blood
circulation, cardiovascular diseases in the mother,
diseases of the kidney and liver, and late pregnancy
toxemia are of great importance in this respect.
In defining this group a distinction has been
drawn only between those clinical forms of mental
retardation whose etiology has been more or less
clearly determined. The forms in this group are as
follows:
1. Mental retardation arising under the influence
of immunopathological factorsincompatibility of the antigenic properties of the maternal and fetal blood with regard to blood
type and rhesus factors
2. Mental retardation associated with Little's
disease

3. Mental retardation caused by the mother

catching measles during pregnancy (embryopathia rubeolaris)
4. Mental retardation caused by other viruses
(influenza, mumps, infectious hepatitis, cytomegalic inclusion disease)
5. Mental retardation caused by toxoplasmosis
and listeriosis
6. Mental retardation associated with congenital
syphilis
7. Clinical forms of mental retardation caused
by hormonal disturbances in the mother and
by toxic factors (exotoxins and endotoxins)
8. Mental retardation caused by hemolytic
disease of the newborn.
Group 3
The third group is comprised of types of mental
retardation caused by harmful factors acting during the perinatal period and the first three years of
the postnatal period. The clinical forms in this
group occur following exposure to various exogenous factors, e.g., birth injury, postnatal injuries, asphyxia during labor, and injuries and intoxications during the first years of life. These
clinical forms of mental retardation are more
complex in structure since, in their clinical and
morphological characteristics, signs of underdevelopment are combined with residual manifestations of the disease concerned.
The following clinical types of mental retardation may be distinguished in this group:
1. Mental retardation due to birth injury and
asphyxia.
2. Mental retardation caused by craniocerebral
injury in the postnatal period (early childhood)
3. Mental retardation caused by general infections during the first three years of life
(influenza, measles, pneumonia, dysentery,
severe forms of dyspepsia)
4. Mental retardation caused by encephalitis,
meningoencephalitis, or meningitis in early
childhood
5. Mental retardation caused by severe disorders
of sensory functions (blindness, deafness)
6. Mental retardation combined with speech defects
7. Mental retardation due to craniostenosis
8. Mental retardation combined with congenital
hydrocephalus

Classification and Mental Retardation: Issues

Arising in the Fifth WHO Seminar on Psychiatric
Diagnosis, Classification, and Statistics
BY G. TARJAN, M.D., J. TIZARD, PH.D., M. RUTTER, M.B., D.P.M.,
M. BEGAB, PH.D., E.M. BROOKE, F. DE LA CRUZ, M.D., T.-Y. LIN, M.D.,
H. MONTENEGRO, M.D., H. STROTZKA, M.D., AND N. SARTORIUS, M.D.

In conjunction with the official report of the

seminar, this paper discusses in more detail
some of the chief issues considered at the
seminar and outlines the reasoning behind the
recommendations. The issues considered include the integration of child psychiatry and
mental retardation, multiaxial classification,
choice of axes, assessment of intellectual retardation, values and limitations of IQ tests,
assessment of social competence, classification of biological factors, application of the
multiaxial scheme to adult patients, and field
trials to test new schemes of classification.
HIS PAPER by the officers of the Fifth
WHO Seminar on Psychiatric Diagnosis, Classification, and Statistics gives an
outline of some of the main issues discussed
at the seminar in order to clarify the reasons
for the decisions listed in the official report of
the seminar. We have tried to faithfully represent the tenor of the discussion at the sem-

inar but, in expanding upon the reasoning

behind the official recommendations, we can
necessarily only speak for ourselves. For the
decisions made at the seminar, readers are
referred to the official report (1).
Integration of Child Psychiatry and
Mental Retardation
Although the classification of mental retardation in adults was discussed at the seminar, most of the discussions concentrated
on the problems associated with mental retardation in children. It became clear early in
the seminar that the issues which arose at an
earlier seminar in this program, the Paris
seminar (Third WHO Seminar on Standardization of Psychiatric Diagnosis, Classification, and Statistics held in Paris in 1967),
with respect to child psychiatry (2), applied
equally to mental retardation. Retarded
children may present psychiatric disorders
that require diagnosis and treatment in the
same way as do children of normal intelligence.
It was found that the current International
Classification of Diseases (ICD-8) (3) has
inadequate provision for the coding of psychiatric disorders in children. Most disorders
tend to be coded under category 308, "behavior disorders of childhood," a vague term that
implies that psychiatric disorders in childhood cannot be further differentiated. This
"rubbish-basket" coding no longer represents
the state of knowledge in the subject (3). The
seminar recommended that the outline provided by the Paris seminar (which gives ten
main categories) be accepted as a provisional
scheme to be tried out internationally in or-

der to arrive at a more definitive scheme

when the ICD is next revised in 1975.
Similarly, the concept of mental retardation is very relevant to the child psychiatrist
who sees many children with educational or
intellectual handicaps. Although in the past
child psychiatrists may not have been so
concerned with the subject of mental retardation, this is less true today. Child psychiatry is developing, and child psychiatrists
see many children with educational or intellectual handicaps. The mentally retarded
child requires expert assessment and treatment with regard to intellectual, emotional, behavioral, social, and medical factors.
Because his handicaps are often multiple,
the retarded child may attend a retardation clinic, a psychiatric unit, or a pediatric
department. In order to make comparisons
between different centers it is essential to
have a classification that encompasses each
of these dimensions and that is equally applicable to different kinds of clinics. The
multiaxial scheme suggested for use in child
psychiatry seemed just as well suited for
mental retardation, and it was recommended
that the scheme be adopted for mental retardation also, with the addition of one
further axis.
Multiaxial Classification
The necessity for a multiaxial or multicategory classification scheme arose at the
Washington seminar (as it did at the Paris
seminar) during the case history exercise
when patients showed a psychiatric disorder as
well as mental retardation, or a physical disorder as well as mental retardation. For example, at the Paris seminar there was a case
of a mentally retarded epileptic girl who
showed, in addition, a psychotic disorder.
Participants at the seminar agreed that it was
appropriate to record three elements (psychosis, mental subnormality, and chronic
brain conditions) but, in fact, most people
recorded only one, with a fairly even split
among the three categories as to which one
was chosen for classification.
Similarly, at the Washington seminar there
was a case of a mentally retarded child with a
severe conduct disorder. More than one-third
of the participants did not record the conduct
disorder in their classification coding in spite
of the fact that they agreed in discussion that

it constituted an important part of the diagnosis. Of course, if only one condition is to be

coded, the selection will depend upon the
special interests of the diagnostician. The patient, however, may receive care or treatment
not only from physicians but from teachers,
industrial instructors, psychotherapists, and
social workers. In order that the person concerned with a retarded child successfully perform his function, he must have not only the
information necessary for his own work, but
also a comprehensive picture of the child's
problems and handicaps. Hence all essential
data need to be coded.
ICD-8 does make some provision for the
combined coding of physical disorder in association with mental retardation through the
use of a fourth digit. However, it is not satisfactory because only a limited number of
conditions are covered and because they are
grouped together. Thus mental retardation
with hypothyroidism and mental retardation
with phenylketonuria both have the same
fourth-digit category. Although it is possible
to code Down's syndrome with mental retardation by using a fourth digit, it is not
possible to thus code, for example, cerebral
palsy with mental retardation.
It was pointed out at the seminar that ICD8 does allow the coding of different elements
by the use of multiple categories. Thus it is
quite possible to classify the retarded, epileptic, psychotic child within categories 311
("mild mental retardation"), 345 ("epilepsy"), and 295.8 ("other," including childhood schizophrenia). However, there are no
rules as to how many categories to use; as the
diagnostic exercise clearly showed, participants varied on how many codings they employed, and when they used only one, they
differed on which one they chose. Furthermore, in a number of medical centers the rule
is to code only one diagnosis per patient.
In discussion it was agreed that in most
cases there is little logic used in selecting the
single coding to be used. It might represent
the referral problem, the interest of the particular clinician, the most "serious" condition
in clinical terms, or the condition that is most
relevant to the administrative action taken
(i.e., admission to a pediatric ward, psychiatric unit, or a hospital for the mentally subnormal). Exactly the same problem arose in
multicategory coding with respect to which
disorder was coded first. Modern computer

techniques allow the analysis of multiple

codings, but in practice frequently only the
first-coded category is analyzedgiving rise
to precisely the same problems as when only
one category is employed.
The fact that a condition is not coded can
lead to multiple contradictory interpretations. It may mean that the condition was not
present, that it was present but not thought
important, or it may just reflect the fact that
it was not coded in spite of being thought
important.
A multiaxial scheme is no more than a
logical development of a multicategory
scheme (such as ICD-8), which introduces
modifications specifically to meet these difficulties. Thus ICD-8 could specify that three
categories must always be coded in order to
ensure that everyone codes the same number
of categories. However, this would leave open
the question of which categories to code, and
clinicians might well decide to record quite
different aspects. A short example easily illustrates this point. If a child is knocked
down by a car and receives severe head injuries resulting in hemiparesis, fits, an IQ of 58,
and the later development of a schizophrenic
state, the three codings could be any combination of categories 343 ("cerebral spastic infantile paralysis"), 345 ("epilepsy"), 311
("mild mental retardation"), E814 (injury resulting from "motor vehicle traffic accident
involving collision with pedestrian"), 293
("psychosis associated with other cerebral
condition"), and possibly 295 ("schizophrenia"). 1 Even if only three conditions were to
be recorded, it would be necessary to provide
rules for precedence in selection.
Of course, a multicategory scheme could
overcome the problem of selecting from several categories by specifying that the three
categories must refer to: 1) clinical psychiatric
syndrome, 2) intellectual level, and 3) medical condition. But here there are two further
problems. In the first place, for purposes of
data processing it would be necessary to ensure that the same disorder was always recorded in the same position among the three

codings selected. Secondly, ICD-8 has no

provision for "no abnormality" with respect
to these three areas. If such a coding is added,
a multiaxial scheme of the type proposed by
the Paris seminar is arrived at.
A multiaxial classification simply regroups
the categories of ICD-8 (with appropriate
modification where necessary) under broad
headings called "axes," provides a "no abnormality" coding in each, and requires that
every case receive some coding on each axis.
In short, a multiaxial scheme is just a reordering of a multicategory scheme with the
addition of simple rules on usage in order to
ensure that everyone interprets the scheme in
the same way.
Unreliability in classification may be due
to several factors, including unreliability in
diagnosis and unreliability in coding. A multiaxial scheme is designed to reduce the errors
in coding and so enable a more valid picture
of morbidity. Serious distortions may stem
from faulty and incomplete recording of
conditions. All the diagnostic exercises in the
seminars in the program on the Standardization of Psychiatric Diagnosis, Classification,
and Statistics showed that even when everyone agreed on diagnosis, there were often serious disagreements on classification,
purely through uncertainty on how to code
when there was a multiple handicap disorder.
It is hoped that a multiaxial scheme will
eliminate this particular problem and so enhance the value of classification for the purpose of unambiguous communication. How
successful it will be will need to be tested in
field trials, some of which are in preparation
and others of which are already being carried
out. The remaining problem of the same
diagnostic term being coded to different
categories can be overcome only by means of
a glossary, a first draft of which has been
prepared (by WHO) and is being tried out.
Choice of Axes
Once it had been decided to recommend
the adoption of a multiaxial scheme, there
had to be a choice of axes. It is obvious that
any given clinical case has many clinically
important aspects, and in order to have a
workable and relatively simple scheme, it was
necessary to restrict the number of axes. Axes
were chosen on the basis of providing unambiguous information of maximum clinical

usefulness in the greatest number of cases.

With mental retardation, one axis had to
pertain to intellectual level, for this has been
shown to be of both medical and educational
or occupational importance. Nearly all individuals with an IQ below 50 have demonstrable brain disease or damage, whereas for
those with an IQ above 50, social factors are
of greater importance in etiology, though an
important minority of cases are due to brain
disorders (4, 5). Thus retarded children with
IQs below 50 come from all social groups,
with a distribution similar to that of the general population. In sharp contrast, retarded
children with IQs above 50 only infrequently
come from professional families; most come
from socially deprived sections of the community.
The IQ level is also of considerable educational importance. Nearly all children with
IQs above 50 can learn to read and1 write and
can attain useful scholastic competence in
other subjects. A few children with IQs from
35 to 50 will gain at least some primitive
reading skills, whereas virtually no children
with IQs below 35 will do so. Similarly, most
adults with IQs above 50 but without other
gross handicaps are employable, whereas few
of those with IQs between 35 and 50 are capable of working in open employment, but
can work in a sheltered environment. Of
those adults with IQs below 35, none is likely
to gain a job outside an institution, and most
will be capable of only the most simple tasks
under detailed supervision (6-8).
It should be emphasized that there is no
qualitative distinction between these various
IQ levels. Rather, the point is that on the
continuum of intellectual ability, a person's
level is of considerable predictive importance.
Many retarded children are under care
largely because of the psychiatric problems
they present, and it was therefore necessary to
record the clinical psychiatric syndrome on
another axis. Retarded individuals can suffer
from any of the syndromes found in those of
normal intelligence, and there is only a very
weak association between intellectual level
and type of clinical psychiatric syndrome.
Accordingly, it was decided to use the same
scheme as that employed for individuals of
normal intelligence. For children, this meant
the scheme suggested by the Paris seminar.
Many cases of mental retardation are associated with brain disease or disorder, and it

was evident that medical classification must

include an axis for this information. Thus a
third axis was called "associated or etiological biological or organic factors." In ICD8 the fourth digit of the mental retardation
codings was intended to meet this need. Unfortunately, it proved unreliable in the case
history exercise, and an alternative was evidently needed. Ideally, this axis should provide a summary classification of the physical
disorders coded elsewhere in the ICD, but the
means for doing so were left to be decided
later. The principles upon which such an axis
might be based are considered in more detail
later in this paper.
Finally, it was recognized that in mental
retardation, as in other types of psychiatric
disorder, psychological and social factors
might be of prime importance in etiology.
Accordingly, it was recommended that there
should be a fourth axis for the coding of these
factors. There is already an E section in ICD8 dealing with external causes (such as accidents and excessive heat or cold), so that this
recommendation does not introduce a new
principle. Nevertheless, there is at present no
available scheme for the classification of
psychosocial factors, and it was thus necessary to recommend that a working party be
formed to develop appropriate categories and
to provide definitions for them.
Assessment of Intellectual R e t a r d a t i o n
Several issues arose in connection with the
assessment of intellectual retardation.
Intellectual Functioning
First it was agreed that mental retardation
referred solely to intellectual functioning and
not to social impairment due to other handicaps (such as sensory defects, physical handicap, emotional disorder, or behavioral disturbance). Although mental subnormality
has sometimes been used as a portmanteau
category to include psychopathy and delinquency even when intelligence is normal or
above normal (9), it was apparent that such
usage would be likely to lead only to diagnostic confusion. It was therefore recommended that mental retardation be diagnosed
only when there is intellectual impairment.
Current State
It was recognized that intelligence is not a

fixed and immutable quality and that in the

present state of knowledge prognostications
about future intellectual development are
necessarily uncertain. Heber has clearly
stated that "a person may meet the criteria of
mental retardation at one age level and not at
another; he may change status as a result of
'real' changes in intellectual functioning; or
he may move from [a] retarded to [a] nonretarded category as a result of a training
programme which has increased his level of
adaptive behaviour" (10, p. 238). There was
general agreement with this view. Classification and diagnosis must be firmly based on
the present and not on crystal-gazing into
the future. The coding on this axis must
therefore deal only with the person's current
state.
Behavioral Criteria
The Paris seminar was quite explicit in
advising that the coding of intellectual level
should be based on the child's current level of
intellectual functioning without regard to its
nature. This ruling was necessary because of
theoretical disagreements about the nature of
certain sorts of retardation. For example,
where there is both psychosis and mental retardation there is no agreement among clinicians on how to decide whether the psychosis
caused the retardation, whether the retardation caused the psychosis, or whether both are
due to a third factor, such as organic brain
disease. To avoid differences in coding due
solely to such theoretical disagreements it is
necessary to specify that the coding of intellectual level be based on the current level of
functioning without regard to views on possible pathogenesis.
A similar problem arises in cases of mental
retardation associated with psychosocial deprivation. Participants at the seminar agreed
that mental retardation may be secondary to
environmental influences and that in such
cases the current functioning may bear little
or no relationship to innate intelligence. By
its very nature, innate intelligence is a concept that cannot be directly measured. The
coding of intellectual retardation should be
based solely on the current level of functioning, and it should be recognized that the coding carries no necessary implications about
innate intelligence. It is a behavioral designation, not a speculation about hypothetical
potentialities.

Use of Intelligence Tests

The issue concerning the assessment of intellectual retardation proved to be more controversial. The WHO Expert Committee reporting on the "Organization of Services for
the Mentally Retarded" (11) had advised that
IQ scores should be used to define the level of
intellectual retardation; this approach was
favored by some participants. The arguments
for this method have been detailed elsewhere (12, 13). However, although it was
agreed that the IQ constituted an important
guide to intellectual level, the seminar participants differed from the WHO Expert
Committee on the weight to be attached to
the IQ.
It was recommended that the clinical considerations of social and adaptive functioning
should always be taken into account and that,
in evaluating the grade of intellectual retardation, note should be taken of the individual's social and cultural background.
Thus a diagnosis of mental retardation implies that the person so diagnosed is both intellectually retarded and socially incompetent. Not all intellectually retarded persons
would be classed as suffering from mental
retardation, but all mentally retarded persons
must show intellectual retardation. Because
the use of intelligence tests proved to be so
controversial, this issue will be discussed
more fully.
Values a n d Limitations of
Intelligence Tests
In Western Europe, North America, and
many other parts of the world where English
and French are widely spoken, intellectual
ability is usually assessed by means of intelligence tests. An intelligence test is made up
of a series of subtests or tasks, each of which
is thought to require intelligence for its successful completion. That is, the items selected
for inclusion in a test battery will be those on
which intelligent or clever people (as judged
by other criteria) tend to be successful,
whereas unintelligent or stupid people of the
same age tend to fail.
Sophisticated statistical methods are used
to select items for inclusion in intelligence
test batteries and, in standardizing an intelligence test, extensive field trials are first carried out on samples of the population to

which the test will later be applied. During

the standardization trials, items that are
found not to discriminate between dull and
bright individuals (as judged by other criteria)
are eliminated, as are items on which one sex
shows superiority over the other. In selecting
items, efforts are made to choose those that
depend as little as possible upon specialized
knowledge (e.g., literacy or computational
facility) for their solution.
Because of the way in which tests are constructed, the most widely used intelligence
tests (such as the various modifications of
Binet's original test and the scales devised by
Wechsler) have been found to give comparable distributions of scores when used in different industrialized countries. The Wechsler
scales have been translated into different
languages, and it has been found that only
minor modifications need to be made in the
wording to make the scales usable in different
countries.
This factand it is a factcoupled with
the demonstration that clinical ratings, unstandardized judgments, and psychophysiological measures of intellectual ability are
highly unreliable and that assessments based
on them correlate poorly with what is ordinarily regarded as intelligence or intelligent
behavior, has led to the widespread use of intelligence tests to assess intellectual ability.
Within a given culture there is no better way
of making comparative judgments of intellectual ability, and it is a mistake to think
that "clinical" procedures produce better estimates. Unfortunately, too few countries
have properly developed and standardized
tests.
Even so, the considerable limitations to intelligence tests are still not sufficiently appreciated, although psychologists have long
been aware of them. They include the following:
/. Error Factors
All measurement is subject to error, due to
a variety of causes. The only satisfactory way
to assess error is to see what happens when
different examiners give the same test (or an
equivalent form of it) to the same individuals
on different occasions. When this is done, IQ
tests are found to have a low number of errors. For example, on the Terman-Merrill
revision of the Stanford-Binet Scale, the testretest correlations obtained range from 0.83

for bright preschool children to 0.97 for dull

school children (14). This level of accuracy is
considerably above that obtained by clinical
assessments, and it is important to note that
IQ tests are most reliable in their assessment
of retarded individuals. In many branches of
medicine relatively little attention has been
paid to the problems of reliability, but where
they have been examined, all clinical examinations have been found to have considerable
error rates (15).
Nevertheless, even this low rate of error
carries implications for classification, as
Shapiro (16) has pointed out. For example, a
school child with an obtained IQ of 69 (indicating mild intellectual retardation) would
have IQ scores between 64 and 74 in 19 tests
out of 20 if he were retested an infinite number of times. The practical result of this is
that a child may score in the intellectually
retarded range one day and not the next.
In order to show the extent of these error
limits, Shapiro recommended that psychologists routinely include a statement on them
in a test report. Thus, for the child with an IQ
of 69, the report might say that his score was
69 plus or minus 5, meaning that in 95 percent of the times a child was tested the limits
of that child's IQ lay within 5 points on either
side of 69. This would indicate that the test
result only showed that the child's intelligence lay somewhere in that range (64 to 74)
and that it was not possible to be more precise
than that. This also applies to clinical judgment of adaptive skills.
2. Errors in Test Construction
In addition to the inherent unreliability of
tests there are systematic errors that arise
from the manner in which tests are standardized. Although tests are expected to have
a mean and median of 100, this is not always
the case in practice. For example, one study
carried out by Dearborn and Rothney(17)
revealed that the median score for nine intelligence tests administered at different times
over a period of years to 320 children ranged
from 94 to 110. There are also differences in
the spread (or standard deviation) of scores
on different tests (on some tests the spread is
uneven, although it should not be), and different tests measure slightly different abilities. For all these reasons it is possible for
someone to score in the mentally retarded
range on one test but not on another test. This

also applies to different types of clinical

judgments.
3. Errors in Administering Tests
An IQ test makes certain assumptions:
for example, that the person can hear the instructions, that he can understand the language, and that he is physically able to make
the responses. If a test that relies on spoken
instructions is given to a deaf child, then the
result will be meaningless. If the psychologist
speaks English and the child understands only Spanish, the IQ score obtained will be
valueless. No competent psychologist will
make such obvious and elementary errors,
but the fact remains that administrative decisions are sometimes (quite wrongly) made
on the basis of IQ tests given in a totally
inappropriate way. This is, of course, a criticism of the usage rather than of the tests
themselves.
4. Cultural Factors in the Content of Tests
Children brought up in different societies
are likely to have different experiences. These
differences in experience may mean that
items in IQ tests may have quite different
implications for them. To take an absurd example, asking a Spanish child to name the
president of the United States is a much more
difficult question than it is to a native New
Yorker. But in more subtle ways, experiences
of language in the home may influence children's responses to items testing verbal skills.
In an attempt to bypass these difficulties, a
few years ago attempts were made to devise
first "culture-free" and then "culture-fair"
teststhat is, in the first case, tests that were
genuinely free from or removed from cultural
context or bias and, in the second case, tests
that only drew upon elements that might be
presumed to exist in any human culture.
Theoretically, this is a dubious procedure,
and it has not worked in practice. Children
from developing countries or from slums and
ghettos in Western societies have often scored
even lower on these tests than they have on
traditional tests.
Anastasi(18) has put the matter succinctly:
To criticise tests because they reveal cultural influences is to miss the essential nature of tests.
Every psychological test measures a sample of
behaviour. Insofar as culture affects behaviour its

influence will and should be reflected in the test.

Moreover if we rule out cultural differentials from
a test we might thereby lower its prognostic
validity. The same cultural differentials that impair an individual's test performance are likely to
handicap him in school work, job performance, or
any other activity we are trying to predict....
Tests cannot compensate for cultural deprivation
by eliminating its effects from their scores. On the
contrary, tests should reveal such effects, so that
appropriate remedial steps can be taken. To conceal the effects of cultural disadvantages by rejecting tests or by trying to devise tests that are insensitive to such effects can only retard progress
toward a genuine solution of social problems. Such
reactions towards tests are equivalent to breaking
a thermometer because it registers a body temperature of 101.
If the diagnostician, being aware of the
dangers in the interpretation of IQ tests, decides to ignore manifest signs of intellectual
retardation in certain of his patients on the
grounds that nearly all people living in their
particular circumstance "behave like simpletons," he will do a serious disservice to
those retarded persons who do not have the
intellectual resources to cope in other than
sheltered circumstances. The needs that led to
the provision of special services for mentally
retarded persons in developed countries do
not cease to exist in developing ones; indeed,
they may actually be greater.
On the other hand, unless the IQ test that is
used has been thoroughly tested out in the
cultural context in which it is to be applied,
the test norms given in the manual cannot be
taken at face value. It would obviously be
wrong to use a test dependent on a particular
set of culture-bound experiences to measure
the intelligence of a child from a different
culture in order to predict his performance in
his own culture. Attempts to do so are likely
to lead to serious underestimates of an individual's intellectual capacity and, on a
wider scale, to give rise to absurdly inflated
estimates of the rates of mental retardation.
This holds for any society, not merely an
impoverished one. Unfortunately, there are
still many societies in the world in which IQ
tests have not been standardized.
However, the difficulty is not solved by
simply being ignored. Psychiatrists and psychologists working with impoverished and illiterate or semiliterate patients therefore
have a responsibility to ensure that before
making judgments, which are essentially

comparative, about the intellectual ability of

their patients, they first provide themselves
with the bases upon which such comparisons
can be made. Where normative data are few,
only the grossest of handicaps can be diagnosed with confidence. Again, it should be
emphasized that these difficulties apply to
clinical judgments in the same way as they do
to IQ scores.
J. Cultural Factors in the Response to Tests
A person's performance on IQ tests depends on motivational and situational factors (18-20) as well as on intellectual capacity. This means that great care must be
taken in ensuring that tests are given under
the best circumstances for the testee and in
making sure that he is interested and involved
in the test. In order to determine that this has
been achieved, it is important to assess the
validity of the IQ score in relation to the
manner of the child's response to the test, his
behavior at home and in other settings, and
all other relevant findings from observation
and investigations. This is of course an intrinsic part of a proper psychological assessment, but clinicians need to beware of
carelessly using tests in a rigid and narrow
fashion.
However, there are more specific examples
of the influence of motivational factors with
regard to the testing of individuals from certain minority subcultures. It appears from
research both in Britain and the United
States (21,22) that attitudes toward racial
differences so influence rapport that white
testers have a subtle deleterious effect on
Negro subjects' scores. This effect probably
only applies in a competitive situation when
Negro subjects know that they are being directly compared with white subjects. It is not
yet clear how general or how marked this effect is, but the fact that it does exist means
that it would be wise to use testers of the same
racial background as the subject whenever
there is the slightest doubt about intellectual
assessment. Similar effects are likely to apply
to an interview situation; care must therefore
also be taken with respect to clinical assessments.
These caveats on the use of intelligence
tests undoubtedly imply that tests cannot
meaningfully be used in a blind fashion. Testers must be alert to the total situation and
how it influences the person's response to the

test. It is necessary to differentiate between

items failed and items not attempted, between items where the person has really applied himself to the problem and items where
there has been only a token effort. Where
there is a marked discrepancy between an IQ
score and a person's performance in the real
life situation, it is essential to analyze the
reasons for the discrepancy.
Nevertheless, given care and attention to
detail, an IQ test remains the best way of
making comparative judgments of intellectual ability within a given culture. Because a
test only assesses behavior, and not potential,
no test can tell us how an individual would
have functioned if he had been brought up in a
different environment. Accordingly, as Vernon (23) has pointed out, cross-cultural
comparisons of "innate" intelligence are futile exercises.
Assessment of Social Competence
Because of these limitations in the use of
IQ tests, the seminar recommended that assessments of intellectual level be made from
a combination of standardized tests and clinical judgments on social and adaptive capacity. Properly applied, this may result
in the most valid measures of current intellectual performance. Nevertheless, it is important to also bear in mind the limitations
inherent in the assessment of social competence. It has already been pointed out that
such measures are subject to the same error
factors that apply to IQ tests, that in the same
way that different IQ tests measure slightly
different intellectual functions, so assessments of social and adaptive capacity will
vary according to which clinical criterion is
used. The same difficult diagnostic problems
arise when assessments have to be made of
persons who do not share or who only
somewhat share the cultural background of
the clinician.
The hope that clinical judgments will be
less subject to social biases than IQ tests has
not been borne out in practice; this means
that this factor should receive particular attention in the training of clinicians. Indeed,
where tested, clinical judgments have generally fared worse than tests. For example, in
Britain, as a result of a proper concern over
the possible social biases of IQ tests as used in
the assessment of children's scholastic abili-

ties, some local authorities have dropped

standardized tests and replaced them with
teachers' judgments for the purpose of
choosing the most able children for selective
schooling. The result has been an increase in
the discrimination against working-class
children in that fewer intelligent children
from poor backgrounds are being chosen for
the academic schools (24). Whether there
would be the same result at the other end of
the intellectual scale is not yet known.
At present there are few standardized tests
of social and adaptive skills, and those that
do exist suffer from important defects for the
purpose of measuring intellectual retardation. For example, the Vineland Social Maturity Scale has been found to be a poor
predictor of scholastic performance, considerably worse than IQ tests (25). Other
scales have been found to be influenced by
deviant behavior (26,27) so that it is not
readily possible to distinguish social incapacity due to intellectual impairment from
that due to mental disorder or illness.
A further difficulty stems from the fact
that criteria for social adaptation are dependent on how exacting the demands happen
to be at that point in time (28). An individual
cannot be considered in isolation from his
social setting; a person's intellectual functioning may be influenced by his environmental circumstances, as may his mood or
affective level. Nevertheless, insofar as social
adaptation is taken as the main diagnostic
feature of any disorder, there is a danger that
fluctuations, for example, in availability of
work, may lead to purely artifactual alterations in the prevalence of that disorder.
Thus, when employment conditions are
bad, more intellectually limited people will
be unable to work, and a diagnosis based on
social adaptation will lead to the conclusion
that the prevalence of mental subnormality
has risen. This is clearly nonsense, but the
argument is important because so long as
social criteria define mental subnormality,
there is a danger that illiberal and unthinking
authorities may cause some people to lose
their liberty by admitting them to a hospital
for the mentally subnormal purely because
employment happens to be scarce at that
moment.
It should be clear that there is no perfect
measure of intellectual retardation; clinical
measures and standardized tests both have

their strengths and weaknesses. The seminar

recognized this and recommended that intellectual level be assessed by thoughtfully applying the fullest possible information on intellectual functioning to the current situation,
with proper regard for the person's social
situation and relevant motivational factors.
Classification of Biological Factors
It has already been stated that the assessment of a mentally retarded person must include both psychological and medical features. It is important to know whether the
retardation is associated with a neurological
and physical condition, since this may influence both treatment and prognosis. The
fourth-digit system of ICD-8 has proven
inadequate for this purpose, and the seminar
recommended that a working party be convened by WHO to consider how this need
might best be met. In the meantime, some
provisional decision had to be reached to
proceed with the field trials (discussed later in
this paper) of the scheme proposed by the
Paris and Washington seminars. It may be
appropriate here to consider the alternatives
that are available.
In general, there are three different approaches that may be followed. Pathogenic
factors or causal influences may be classified,
physical handicaps may be classified without
regard to their causation, or classification
may be principally concerned with recognizable medical conditions. The decision as to
which is preferable depends on considerations
of what is practical and on which method
gives the most useful clinical information and
the greatest predictive power.
It is sometimes thought that an ideal classification should always be based on etiological mechanisms. However, this depends on
the purpose for which the classification is required. For example, a classification of fractures based on whether the bone was broken
by the patient's falling from a tree or being
hit by a car is useful when preparing statistics
designed to foster the prevention of accidents,
but from the viewpoint of treatment it is less
useful than one based on the nature and extent of bony- and soft-tissue damage (i.e.,
simple, compound, complex, and comminuted fractures). Which method is best for
mental retardation is therefore an empirical
question, subject only to the condition that a

diagnosis should, above all, be descriptive.

ICD-8's fourth-digit system for use with
categories of mental retardation is based on
pathogenesis. Thus a fourth-digit coding of .3
in mental retardation means an association
with gross (postnatal) brain disease, and .5
denotes an association with chromosomal
abnormalities. There are three major problems with this system. First, most cases of
mental retardation are of unknown etiology,
and a classification based on causes tends to
do no more than express the clinician's
theoretical predilections. Berg and Kirman's
survey of hospitalized mentally retarded patients (29) showed that the only large group
of known etiology were patients suffering
from Down's syndrome (23 percent of the
patients). There were another ten percent with
known disorders and four percent with a
probable disorder; but for 32 percent of the
patients it was only possible to surmise the
cause, and for 31 percent the cause was
unknown.
The second difficulty is that in many cases
of mental retardation there is multiple
causation, and certain pathogenic influences,
by their very nature, group together. For example, in the fourth-digit system .6 (prematurity) is frequently due to .4 (prenatal influences), which is in turn associated with .1
(perinatal trauma). It is now known that the
cerebral damage in premature infants is often
caused by severe hypoglycemia in the postnatal period, which would further necessitate
a coding of .2 (disorder of metabolism,
growth, or nutrition). It is by no means clear
which of these four codings should have
precedence. It was in part this kind of confused overlap that led to the unreliability of
the fourth-digit coding in the Washington
seminar case history exercise.
The third major difficulty associated with a
pathogenic classification is that different
disorders may be due to the same cause, and
thus the classification will not reflect important diagnostic distinctions. For example, the
fourth-digit system cannot tell one whether
mental retardation is associated with cerebral
palsy because often both are due to the same
perinatal causes.
An alternative system is to totally omit
questions of either etiology or physical
disease and instead code the accompanying
handicap. Thus one might code convulsive
disorder, motor defect, visual handicap, and

sensory impairment. Such a system would

undoubtedly provide clinically useful information. The physical handicaps of a retarded person may be of crucial importance
in planning services to meet his needs. On the
other hand, in using such a system it would
not be at all easy to differentiate important
conditions such as Down's syndrome. Furthermore, most patients have multiple handicaps, and the use of five or six codings on this
axis alone would be tedious and complicated
to handle statistically. A classification of
handicaps may be the must useful system, in
conjunction with other categories, for research or clinical purposes, but it does not
seem suitable as the prime principle of coding
on the axis for associated or etiological biological factors.
The third systemthat of coding medical
conditionsis more of a compromise and in
some ways less pure and less logical than the
other two, but in practice it appears to be the
most satisfactory system. In this system, for
example, cerebral palsy would be coded rather than the fact that it is thought to be due
to perinatal damage. This means that information on how the cerebral palsy was
caused is lost, but information about the
physical and neurological handicap is retained in more precise form. Cerebral palsy
may be due to perinatal damage or to a variety of postnatal insults occurring during infancy (e.g., encephalitis, head injury, cerebral
abscess). However, for most purposes it is
more important to know that a child is currently hemiparetic than to know what caused
the disorder many years ago.
In cases where there is a one-to-one relationship between cause and condition the
coding will of course give the same information as a pathogenic classification. This is
true of all the well-defined diseases of known
etiology, such as Down's syndrome, phenylketonuria, and hepatolenticular degeneration. The difference chiefly occurs with
conditions of variable and often unknown
etiology, such as cerebral palsy and epilepsy.
However, since these are common disorders
that are better coded on the "condition" system, the advantages probably lie with this
approach. Because of this, it was chosen for
the field studies testing out the classification
scheme proposed by the Paris and Washington seminars. However, as with each of the
three methods of classifying associated or

etiological biological factors, there are problems in deciding how to deal with some conditions. Time and testing will show whether
this proves to be the best system.
Application of the Multiaxial
Classification to Adult Patients
Little time was spent at the seminar in discussing classification of mentally retarded
adults. In the same way that the classification
of child psychiatric disorders must be compatible with the scheme used to classify adult
psychiatric disorders, so must the classification of mental retardation be developed in
such a way as to apply to all age groups.
In general, it was agreed that the multiaxial scheme devised for children should be
equally suitable for adult patients. It is just as
necessary in adults as in children to classify
the degree of intellectual impairment, associated biological condition, associated psychosocial factors, and accompanying mental
disorder or clinical psychiatric syndrome.
However, there is a less close relationship
between intellectual level and school attainment. This means that there will be a
more tenuous relationship between IQ and
social handicap in adults and, furthermore,
that many retarded individuals who were
handicapped in childhood will not be retarded
as adults (30). These are matters of detail,
however, and the principles of classification
are the same at all age levels. Whether in fact
the scheme proposed works as well for adults
as for children is an empirical question that
needs to be answered by field trials.
Field Trials
In the past the production and revision of
schemes of classification have all too often
been an armchair exercise, with changes
made largely for diplomatic rather than
scientific reasons. The current series of WHO
seminars is an exciting new endeavor in
which, for the first time, there is a systematic
attempt to assess the strengths and weaknesses of the existing ICD classification by
means of carefully planned case history and
videotape diagnostic studies. These have been
invaluable in highlighting where and why
there were difficulties in classification, and
they have clearly shown which parts of the
classification need revision or deletion be-

cause of their unreliability.

As a result of this preparation there is a
reasonable chance that any proposed changes
in the classification will be an improvement
over what existed before. Nevertheless, the
proposal of changes or modifications marks
the beginningnot the endof the process of
revising the ICD. It was agreed at the outset
that any scheme agreed upon at a seminar
will be tested through national and international exercises for further refinement and
revision before a final recommendation is
made for the revision of the ICD in 1975.
Accordingly, field trials of the multiaxial
scheme proposed by the Paris and Washington seminars are now in progress in the
United Kingdom, France, the United States,
and Scandinavia. The results of these trials
will be reported to future seminars so that
findings regarding any one part of the classification may be taken into account in considering other parts of the classification.
Summary
In conjunction with the official report of
the Fifth WHO Seminar on Psychiatric
Diagnosis, Classification, and Statistics (2),
this paper discusses in more detail some of
the chief issues considered at the seminar and
outlines the reasoning behind the recommendations. The issues considered include the
integration of child psychiatry and mental
retardation, multiaxial classifications, choice
of axes, assessment of intellectual retardation, values and limitations of IQ tests, assessment of social competence, classification
of biological factors, application of the multiaxial scheme to adult patients, and field
trials to test new schemes of classification.
REFERENCES
1. Fifth WHO Seminar on Psychiatric Diagnosis,
Classification, and Statistics. Amer J Psychiat 128
(May suppl):3-14, 1972
2. Rutter M, Lebovici S, Eisenberg L, et al: A tri-axial
classification of mental disorders in childhood. An
international study. J Child Psychol Psychiat 10:
41-61, 1969
3. World Health Organization: International Classification of Diseases, 8th revision. Geneva, WHO, 1968
4. Tizard J: discussion of Gruenberg EM: Epidemiology of mental illness. Int J Psychiat 2:131 -134, 1966
5. Kushlik A: Social problems of mental subnormality,
in Foundation of Child Psychiatry. Edited by Miller
E. London, Pergamon, 1968, pp 369-412
6. Clarke ADB, Clarke AM: The abilities and traina-

7.
8.
9.

10.

11.
12.
13.

14.

15.
16.

17.

bility of imbeciles, in Mental Deficiency: The

Changing Outlook. Edited by Clarke AM, Clarke
ADB. London, Methuen, 1965, pp 356-384
Gunzberg HC: Vocational and social rehabilitation
of the subnormal. Ibid, pp 385-416
Tizard J: Longitudinal and follow-up studies. Ibid,
pp 482-509
Shapiro A: Delinquent and disturbed behaviour
within the field of mental deficiency, in The Mentally Abnormal Offender. Edited by de Reuck AVS,
Porter R. Boston, Little, Brown and Co, 1968, pp
76-90
Heber R: The concept of mental retardation: definition and classification, in Proceedings of the Conference on the Scientific Study of Mental Deficiency,
London, 1960. Edited by Richards BW. Dagenham,
England, May & Baker, 1962, pp 236-242
Organization of services for the mentally retarded.
WHO Techn Rep Ser 392, 1968
Rutter M, Tizard J, Whitmore K: Education, Health
and Behaviour. London, Longmans, 1970
Rutter M: Psychiatric disorder and intellectual retardation in childhood, in Mental Retardation: An
Annual Review, III. Edited by Wortis J. New York,
Grune & Stratton, 1971, pp 186-221
McNemar Q: Revision of the Stanford-Binet Scale.
An Analysis of the Standardization Data. Boston,
Houghton Mifflin, 1942
Witts LJ (ed): Medical Surveys and Clinical Trials.
New York, Oxford University Press, 1959
Shapiro M: Intensive assessment of the single case,
in The Psychological Assessment of Mental and
Physical Handicaps. Edited by Mittler P. London,
Methuen, 1970, pp 645-666
Dearborn WF, Rothney JWM: Predicting the

18.
19.
20.

21.
22.
23.
24.
25.
26.
27.
28.
29.
30.

Child's Development. Cambridge, Mass, Sci-Art

Publishers, 1941
Anastasi A: Psychological Testing, 3rd ed. New
York, Macmillan Co, 1968
Wickes TA:- Examiner influence in a testing situation. J Consult Psychol 20:23-26, 1956
Zigler E, Butterfield EC: Motivational aspects of
changes in IQ test performance of culturally deprived nursery school children. Child Develop
39:1-26, 1968
Deutsch M, Katz I, Jensen A (eds): Social Class,
Race and Psychological Development. New York,
Holt, Rinehart and Winston, 1968
Watson P: How race affects IQ. New Society
16:103-104, July 1970
Vernon PE: Intelligence and Cultural Environment.
London, Methuen, 1969
Floud J, Halsey AH: Intelligence tests, social class
and selection for secondary schools. Brit J Sociol
8:33-39, 1957
Pringle MLK: Social Learning and Its Measurement. London, Longmans, 1966
Nihira K: Factorial dimensions of adaptive behavior
in mentally retarded children and adolescents. Amer
J Ment Defic 74:130-141, 1969
Nihira K: Factorial dimensions of adaptive behavior
in adult retardates. Amer J Ment Defic 73:868-878,
1969
Wootton B: Social Science and Social Pathology.
London, Allen & Unwin, 1959
Berg JM, Kirman BH: Some aetiological problems
in mental deficiency. Brit Med J 2:848-852, 1959
Gruenberg E: Epidemiology, in Mental Retardation:
A Review of Research. Edited by Stevens HA, Heber
R. Chicago, University of Chicago Press, 1964, pp
259-306