Psychiatric Diagnosis Across Cultural Boundaries

Joseph Westermeyer, M.D., Ph.D.

share even the language, much less the culture or world

Diagnosis across cultural boundaries has become a view, of the clinician.
practical rather than an esoteric matter as migration, Diagnosis involves more than a mere Linnaean
the number of effective psychiatric therapies, and classification of homogeneous entities. In the medical
access to psychiatric care have increased. Cross- context it serves at least one purpose: predicting the
cultural diagnosis involves such theoretical course or the probabilities of outcome from alternate
considerations as diagnostic categories, courses. It satisfies a human need in bringing discom-
pathoplasticity of psychiatric disorder, so-called forting, sometimes terrifying, uncertainty into the
culture-bound syndromes, “emic” (intracultural) realm of the known. Only sometimes does diagnosis
versus “etic” (cross-cultural) conceptual frameworks, meet a second goal: to recommend a course of treat-
and different reporting of symptoms and expression ment. This involves, then, a second prognosis regard-
of signs from one cultural group to another. ing the expected course or courses following treat-
Important clinical issues include distinguishing ment.
cultural belief systems from delusions and
understanding the special problems of minority,
migrant, and refugee patients. THEORETICAL CONSIDERATIONS
(Am J Psychiatry 142:798-805, 1985)
Diagnostic Classification in Various Countries

T he term
meanings.
“cross-cultural
It can refer
diagnosis”
to diagnostic
has different
schemata
Diagnostic
and large ways.
schemata
Some
vary among
examples
countries
of national
in small
diagnostic
across cultures: this is the “classification” meaning. To categories include the following:
what extent are these schemata identical, merely over- 1. Bouffees d#{233}lirantes, a type of acute psychosis
lapping, or essentially different? Cross-cultural diagno- described by the French, consists of a transient psycho-
sis also implies the ability of a clinician from one sis with elements of trance or dream state (1, 2).
culture to make a diagnosis for a patient from another 2. The definition of schizophrenia varies from the
culture: this is the “diagnostic method” meaning. Can tight, narrow syndrome preferred in Western Europe,
the clinician take the signs and symptoms of the Asia, Latin America, Africa, and Australia (3) to the
culturally foreign patient and arrive at a diagnosis? wide spectrum, more inclusive syndrome described in
Other meanings can be discerned, such as cross- the Soviet Union (4) and the United States (5-7). DSM-
cultural comparability of folk diagnoses or interrater III criteria have shifted toward the more specific
reliability of diagnosticians from two different cul- criteria used elsewhere.
tures. 3. Scandinavian psychiatrists have long favored a
Cross-cultural diagnosis concerns the ability of the category of reactive psychosis distinct from schizo-
classification consistently to reflect clinical phenomena phrenia and affective psychosis. It consists of acute
so well that even cultural differences in mode of onset associated with severe precipitating stress, a
complaint and in lay or folk meaning attached to better premorbid adjustment, and favorable outcome
symptoms do not mislead the average competent clini- (8). This category is also approximated by the DSM-
cian away from the diagnosis. To be effective, cross- III diagnosis of schizophreniform psychosis.
cultural techniques, skills, and conceptual frameworks 4. Japanese psychiatrists recognize a syndrome
must be available for evaluating a patient who may not marked by obsessions, perfectionism, ambivalence,
social withdrawal, neurasthenia, and hypochondriasis.
They refer to it as shinkeishitsu (9).
Received Oct. 17, 1983; revised March 19 and Aug. 13, 1984; S. During the 1950s to 1970s, Chinese psychiatrists
accepted Aug. 20, 1984. From the Department of Psychiatry, preferred the generic term “neurasthenia” for all types
University of Minnesota. Address reprint requests to Dr. Wester- of neurotic disorders rather than more specific catego-
meyer, Department of Psychiatry, University Hospitals, University ries (10).
of Minnesota, Box 393, Mayo Memorial Bldg., 420 Delaware St.,
S.E., Minneapolis, MN 55455. 6. In Spain and Germany, “involutional para-
The author thanks Jerry Kroll for his critical comments. phrenia” is a midlife condition with paranoid features.
Copyright 0 1985 American Psychiatric Association. In these countries, “paraphrenia” is distinct from

798 Am J Psychiatry 142:7, July 1985

 JOSEPH WESTERMEYER

schizophrenia and depression while containing ele- syndromes in one culture may mask what is termed
ments of both (11). “depression” in other cultures.
Are these cross-national differences due to true Pathoplasticity may also be manifest in alcohol and
national differences in psychopathology or merely to drug abuse. In societies with high rates of substance
national preferences that have developed over time? abuse, the rate of neurotic-type disorders is corre-
Often, national preferences and historical factors ap- spondingly low (26). This finding suggests that alco-
pear to account for these differences, but national holism and drug abuse may be an alternative to
variations in psychopathological expression may also neurotic disorders. Other data supporting this inter-
exist. pretation include the genetic and family links between
In contrast to these idiosyncratic national variations, depression and substance abuse, the correlation of
highly similar diagnostic categories are also used by parental loss during childhood with both types of
psychiatrists around the world. These categories in- disorder, the inverse sex distribution of these disor-
dude manic-depressive disorder and various forms of ders, with more depressed females and more sub-
neurosis, alcoholism, drug dependence, mental retar- stance-abusing men, and similar age at onset of both
dation, organic brain syndromes, personality disorder, disorders (27).
and various adjustment disorders (12-15).
Diagnostic criteria can and do change over time. Culture-Bound Syndromes
Perhaps most notable over the last decade has been the
movement in the United States from emphasis on Certain trance-like or behavioral disturbances occur
psychodynamic and psychoanalytic concepts and fairly with unusual frequency in certain societies. A few
loose diagnostic criteria before the 1970s to more examples of these many syndromes include the follow-
specific diagnostic criteria since the 1970s (16, 17). ing:
This change has led to a decrease in the diagnosis of 1. Latah in Southeast Asian females: minimal stim-
schizophrenia (17), although as late as 1980, older uli elicit an exaggerated startle response, often with
American clinicians were still using broad criteria for swearing; also reported among the Ainu of Japan, the
this diagnosis (18). Attitudinal, political, historical, Bantu of Africa, and French-Canadians (28).
and perhaps even economic factors can also influence 2. Anthropophobia among Japanese, especially
diagnostic criteria and diagnostic practices (4, 19-21). males: involves easy blushing, anxiety with face-to-
face contact, and fear of rejection (29).
Pathoplasticity of Psychiatric Disorder 3. Koro, especially in Asian males: consists of the
fear that the penis will withdraw into the abdomen,
As has been repeatedly demonstrated, the concomi- causing death; also reported infrequently in Europe
tants of schizophrenia vary widely from one culture to and America (30, 31).
another and even among ethnic groups in a single 4. Grisi siknis among the Miskito of Nicaragua:
country. Core elements of the disordered perceptions, involves headache, anxiety, irrational anger toward
thought processes, and behavior are highly consistent, people nearby, aimless running, and falling down (32).
however. Differences mainly involve the content, Se- S. Amok among Southeast Asian males: involves
verity, or relative frequency of such symptoms as sudden mass assault, usually including homicide and
withdrawal, volubility, agitation, compliance, and sometimes the death of the perpetrator (33).
paranoia (22). 6. Bulimia in North American Euroamericans,
Another type of pathoplasticity involves the nonpsy- mostly females: food binging is followed by self-
chotic disorders. For example, the relative distribution induced vomiting; sometimes associated with other
of depression, conversion reaction, anxiety, somato- conditions such as depression, anorexia, or substance
form disorder, and obsession-compulsion differs from abuse.
one culture to another (23). 7. “Falling out” among black Americans and black
A manifestation of cross-cultural pathoplasticity is Caribbeans: consists of sudden collapse and paralysis
the better outcome of schizophrenia in developing and inability to see or speak; hearing and understand-
countries than in developed countries (24). One inter- ing are intact (34).
pretation of these data is that patients in developing Most students of these syndromes agree that they
countries have better outcomes due to social factors are not uniquely “culture bound”; that is, they occur
(e.g., more stable social networks, better work oppor- in a variety of cultures including quite dissimilar ones
tunities). It may also be that acute, self-limited psycho- separated by considerable distances. In addition, the
sis (so-called hysterical psychosis, bouffees d#{233}lirantes, subsegments of these syndromes are not unique but
acute reactive psychosis, or schizophreniform psycho- include such common manifestations as fear, anxiety,
sis)-which may be mislabeled as schizophrenia-is amnesia, aimless escape, psychophysiological symp-
more common in these countries. toms, social withdrawal, behavioral deviance, and
It seems likely that certain psychopathological states nondirected violence. Nonetheless, it is remarkable
may change diagnostic category entirely in different that these syndromes do occur with much greater
cultures. Lesse (25) has made the point that syndromes frequency in some societies than in others. They often
classified as behavioral problems or psychosomatic wax and wane over time and can occur in sudden

Am J Psychiatry 142:7, July 1985 799

PSYCHIATRIC DIAGNOSIS ACROSS CULTURAL BOUNDARIES

widespread “epidemics,” such as outbreaks of koro, etic?” Stated differently, to what extent do depressed
bulimia, and amok (35). patients from various cultures express symptoms dif-
These syndromes do not accompany a single psychi- ferently, and to what extent do psychiatrists from
atric diagnosis. Instead, associated psychiatric disor- different cultures recognize the same depressive syn-
ders include personality disorder, “neurasthenia,” cri- drome?
sis or adjustment disorders, organic brain syndromes, Recently, attempts have been made to develop emic
drug-induced delirium, major depression, mania, rating scales for depression among Vietnamese refu-
schizophreniform psychosis, and schizophrenia. gees in the United States (41) and for somatic com-
Some students of these phenomena (mostly social plaints among Nigerians (42). Creation of such scales
scientists) stress the cultural uniqueness of these syn- may contribute to our better understanding of the way
dromes and insist that they can or should be consid- such disorders are reported and/or perceived across
ered only within their own sociocultural context (36). cultures. Some investigators believe that an etic version
Others (usually clinicians) argue that we can learn by of depression across cultures has not been demonstrat-
studying and comparing the biopsychosociocukural ed (43). So far, it has not yet been demonstrated that
concomitants of these disorders wherever they occur emic depression scales have an inherent advantage
(28). This controversy appears to be related to the over well-translated etic depression scales for either
research methods at the command of the various clinical or research purposes, although such research is
investigators. Some social scientists object to psycho- currently underway. It will also be of interest to assess
logical or biophysiological approaches that point up versions of such emic scales translated into other
similarities among these and other disorders, whereas languages, since they hinge mostly on experiences,
physiologists, psychiatrists, and psychologists do use symptoms, feeling states, and attitudes that are more
these research approaches. universally human than specific to only one culture.

Emic and Etic Perspectives Differential Reporting of Symptoms and Expression

of Signs
The terms “emic” and “etic” have been borrowed
from the linguistic concepts “phonemic” and “phonet- There is considerable overlap among
groups ethnic
ic.” Emic refers to socially unique, intracultural per- with regard to presenting symptoms. in a For example,
spectives, such as susto or “the fallen fontanel syn- Connecticut patient registry, sleeping problems and
drome” among Mexicans. Etic refers to universal, eating problems were the first and second most com-
cross-cultural concepts such as febrile dehydration or mon symptoms among white, black, and Puerto Rican
hypertrophic pyloric stenosis that can occur in any patients (44). In addition, the next three most frequent
cultural group (and which, for example, could account symptoms (other physical problems, suicidal thoughts,
for susto in an infant). Cross-cultural diagnosis (an etic and relationship disturbance with a child) were report-
activity) and so-called culture bound syndromes (sup- ed in the three ethnic groups, but with different relative
posedly emic phenomena) are closely tied to the emic- frequencies. Thus, the similarities were impressive, but
etic concept. there were evident differences as well.
Psychiatric diagnosis has involved methods and clas- Leff (45) demonstrated a highly significant differ-
sification systems that basically are etic, even though ence with regard to terms differentiating affect between
certain emic considerations (such as the content of six Indo-European languages and two non-Indo-Euro-
delusions) must be considered. Even psychometrics pean languages. He reports that the Indo-European
(such as the MMPI) and self-rating scales (such as the patients showed greater affective differentiation than
Zung Self-Rating Depression Scale) have been translat- did the others. His findings probably relate to the fact
ed into numerous languages and restandardized using that he chose symptoms such as “anxiety” and “irrita-
the original concepts and symptoms. Concepts reflect- bility” that are more elaborated in the Indo-European
ed in such tests and scales may be culture bound so languages, while he ignored affective concepts such as
that, even with a grammatically accurate translation, those involving degrees of sadness, loss, and frustra-
the translated instrument may not be strictly compara- tion that are better developed in other languages.
ble to the original. Symptoms may also be differently Despite my quibble (because of my familiarity with the
attended to or reported among various cultures. A Lao and Hmong languages), Leff has shown that
large and growing literature does indicate the utility of languages do differentially favor this or that affective
these instruments across cultures, but with enough expression-a valuable contribution.
minor variability so that small differences observed Cultural differentiation extends to the expression of
between two cultural groups may be due to the affect in facial display (physiognomy) (46, 47). Within
instrument and not to true cultural differences (37- cultures, affect can be discerned with a high degree of
39). Use of the Hamilton Rating Scale for Depression interrater reliability, and this reliability persists to a
in Italy has demonstrated that scales rated by psychia- considerable extent across cultures. However, at least
trists may have very high intercultural validity (40). one study indicates that strong emotion is modulated
Thus the diagnostic issue may not be “emic or etic?” or expressed differently among Japanese than among
but rather “to what extent emic and to what extent Americans (48).

800 Am J Psychiatry 142:7, July 1985

 JOSEPH WESTERMEYER

The endocrine, physiognomic, and central nervous should belong to the diagnostic group under consider-
systems make up the biological infrastructure for ation (59).
affective, behavioral, and cognitive functions. Little Application of the kappa statistic to psychiatric
research attention has been devoted to this axis across diagnosis has demonstrated that psychiatrists agree
cultures. However, the biological infrastructure may with each other about as often as electrocardio-
be a factor influencing cultural differences. Sociocul- graphers agree regarding abnormal ECGs or radiolo-
tural factors can indeed lead to malnutrition and gists agree on abnormal intravenous pyelograms. Di-
growth failure (49), and even neonates can demon- agnostic reliability among psychiatrists is high for
strate differences in behavior across ethnic and cultural organic psychiatric syndromes (K=.7-.8), intermediate
boundaries (SO, 51). Adult disorders with uneven for schizophrenia (K=.S-.6), and low for depression
geographic distribution (e.g., cerebral malaria, filaria- and affective disorder (K=.3-.4) (65). These interrater
sis) can produce organic brain disorder (52). Thus, reliability scores can be improved considerably by
there may be culturally or geographically determined using specific diagnostic criteria (66), but this results in
biological differences that influence the expression of a number of undiagnosed cases (67, 68). Discordant
signs or symptoms. rating is greatest for cases in which the symptoms and
Despite these cultural differences in signs and symp- signs cut across diagnostic categories, such as an
toms, similarities exist as well. Orley and Wing (53) admixture of schizophrenic findings with depression
found a level of symptoms in African villages that (69, 70).
was twice as high as the level in London, but the actu- Soon after the development of the kappa statistic,
al morbidity rates did not vary so greatly. Marks (54) comparisons of cross-cultural diagnosis began to ap-
observed that African and English university students pear. Numerous methods have been used, including
had similar prevalence rates for psychiatric morbid- actual interviews of the patients (71), films and video-
ity. tapes of patient interviews (37, 72), and written case
Thus far, few cross-cultural data exist regarding reports (70). Interestingly, regardless of method, these
child psychiatric patients. Available clinical data sug- cross-cultural studies have given kappa scores compa-
gest that cultural factors may account for greater rable to intracultural reliability scores.
differences among child than among adult psychiatric The first studies of cross-cultural psychiatric diagno-
patients (55). sis involved comparisons between English and Amen-
can psychiatrists (71, 73-75). Greater international
representation appeared in later studies, with Japan,
EVALUATION OF CROSS-CULTURAL DIAGNOSIS Sweden, and other countries represented (76, 77).
Later, similar studies were conducted under the aegis
Interrater reliability is an important means for eval- of the World Health Organization (24, 78). These
uating cross-cultural diagnosis. However, its utility is studies showed certain diagnostic biases, such as the
limited to a single event in time: the ability of two tendency for American psychiatrists to diagnose
diagnosticians to agree with each other. A more impor- schizophrenia in cases that psychiatrists elsewhere
tant characteristic of cross-cultural diagnosis is valid- recognized as affective or brief reactive psychosis (6).
ity. The latter depends on observing the course and the Psychiatrists from industrialized countries did well in
effects of treatment over time. Reliability often accom- their interrater scores with patients from nine industri-
panies validity, but the reverse is not always true. alized societies, but they had poorer reliability scores
Thus, we cannot rest content on establishing the with patients from four developing countries. Psychia-
reliability of cross-cultural diagnosis; validity must tnists from the developing countries did well with
also be demonstrated. patients from both developing and industrialized coun-
tries, probably because they had experience with pa-
Diagnostic Reliability tients from both types of societies as a result of their
having trained in industrialized countries (72).
Concerns about diagnostic reliability in psychiatry In line with these data, clinicians in one cross-
(i.e., the probability that two clinicians will agree with cultural study (70) had the highest confidence in their
each other’s diagnosis) go back more than half a organic diagnoses, intermediate confidence in diag-
century (56). Much work on reliability of psychiatric noses of schizophrenia, and low confidence in diag-
diagnosis was done during the 1940s and l9SOs (57- noses of affective psychosis.
60), but it was not until the late 1960s and early 1970s
that quantitative measurement of diagnostic reliability Diagnostic Validity
became available with the kappa statistic (61-64).
Perfect kappa agreement is 1.0 and random agreement Cross-cultural diagnosis can be assessed by observ-
is 0.0. The kappa scoring method usually measures ing treatment variables. A pilot study of 15 Hmong
only one category at a time, such as “schizophrenia” refugee patients in the United States was undertaken
and “not schizophrenia,” although more categories for this purpose (79). All 15 outpatients had major
can be compared. For more robust and consistent depression according to DSM-III criteria, including six
results in using the kappa statistic, about half the cases with simple depression, four with melancholic depres-

AmJPsychiatry 142:7,July 1985 801

PSYCHIATRIC DIAGNOSIS ACROSS CULTURAL BOUNDARIES

sion, and five with psychotic depression. Those pa- Self-Rating Depression Scale. After lengthy discussion
tients with simple depression were treated over an with Hmong co-investigators and a pilot study, we had
average of 2.8 visits and none received pharmacother- decided that the translation of an item on sexual
apy. Melancholic patients had a mean of 6.0 visits, and function was insulting and hence unacceptable to
all received tricyclic antidepressants. Psychotic de- Hmong currently without a sex partner. In order to
pressed patients made 9.8 visits on the average, and devise a culturally sensitive item that still tapped into
combined tricyclic-neuroleptic therapy was prescribed this important topical area, we created two questions
in all cases. for this one item: one for those with and one for those
Validity of diagnosis can also be assessed by study- without sexual partners. We believed our translation
ing outcome. It has been suggested that episodic heavy was culture fair and reduced test bias, and our review-
alcohol consumption, or binge drinking, among Amer- er believed the opposite.
ican Indians does not constitute actual alcoholism.
However, a 10-year follow-up study of 45 American Folk Diagnosis Versus Psychiatric Diagnosis
Indian subjects demonstrated outcomes similar to
those of alcoholic populations rather than those of Several investigators have observed the overlap be-
normal populations (80). There was a high mortality tween folk categories and psychiatric categories. Such
from alcohol-related causes, especially among younger diagnostic categories include mental retardation, epi-
subjects (N=9, 20%), and those with continued heavy lepsy, neurosis, and psychosis (84-89).
drinking had deteriorated in health and/or social status At least one reliability study of two major categories
(N= 19, 42%). Those who were improved in health (psychosis and neurosis) showed good agreement be-
and/or social status (N=7, 16%) had been abstinent tween some psychiatrists and folk diagnosticians. Two
from alcohol for periods of 3 to 10 years (their mean “culturally naive” clinicians (i.e., not familiar with the
abstinence in the past decade was 72 months). These culture studied) had low kappa scores of .21 and .52
data indicate that the cross-cultural diagnosis of alco- when their clinical diagnoses were compared with
holism in this sample was a valid one. local folk diagnoses. Three “culturally experienced”
Another potential means for assessing diagnostic clinicians (i.e., familiar with the culture area of the
validity would be to measure treatment response vis-#{224}- subjects) had high kappa scores of .63, .63, and .68
vis drug dose and blood level of medication. However, when their diagnoses were compared with the local
limited work to date reveals that the ratio of oral dose folk diagnoses (90).
to blood level is quite inconsistent from one cultural
group to another (81). Therapeutic dose levels and Intelligence Testing for Mental Retardation
even EEG responses to dose levels also show much
variability across cultures (82). The meaning of these Diagnosis of moderate or severe mental retardation
data is not clear. Are there racial or nutritional factors across cultural and/or linguistic boundaries poses little
that modify drug responsiveness? Or are there psycho- challenge to the experienced clinician with a modicum
social factors that facilitate therapeutic response at of cultural sensitivity. This is not true with mild and
lower tissue levels? Are both types offactors operative? borderline cases of retardation, however. I have en-
Are there yet other explanations? countered foreign-born and American Indian individ-
uals of average to low-average intelligence who have
been labeled as mentally retarded by psychologists and
CURRENT AND FUTURE RESEARCH ISSUES educators using tests standardized to middle-class,
English-speaking Euroamericans. And, conversely, I
Cultural Bias of the Researcher have encountered several Indochinese patients with
borderline, mild, and moderate retardation who had
Cultural bias can quietly invade the thinking of even been placed in educational settings well beyond their
the most objective empiricist. A renowned example is abilities when intelligence testing was not undertaken.
the nineteenth-century physician, Morton, who mea- These misplaced students were then subsequently re-
sured aboriginal skulls in North America (83). His ferred to me for conduct disorders in the classroom,
own published quantitative data show no significant depression, and/or academic underachievement. Con-
differences in the cranial capacity among races, but on sequently, cross-cultural intelligence testing is increas-
the basis of his apparently unconscious finagling of the ingly important for prevention as well as assessment of
data analysis, he erroneously concluded that Euro- mental disorders.
americans have greater cranial capacity than do Native Unfortunately, intelligence has been reified, perhaps
Americans. even deified, as a concrete entity, rather than viewed as
If this type of bias can infiltrate even measurement a conceptual entity with strong cultural aspects. New
of the cranium, we must be ever vigilant against its thought on this subject may help to dissipate old
incursion into psychiatric research. Of course, it is not notions (91, 92). At this point it seems clear that
always a simple matter to discern bias. Recently, a intelligence testing across cultural boundaries is a
reviewer for research grants indicated that I had highly specialized task that should be conducted by a
introduced a cultural bias into translation of the Zung trained, experienced cultural psychologist.

802 AmJPsychiatry 142:7,July 1985

 JOSEPH WESTERMEYER

Cross- Cultural Personality Assessment cal rather than theoretical or abstract. The first of
these is the setting of a prognosis. Even in classical
Anyone who has plunged into another language, Greek and Roman times, before powerful treatment
culture, and social network has recognized the full methods were available, skilled clinicians were con-
gamut of personality types manifest in the new culture. sulted for their prognostic skills. The second purpose
To be sure, at first blush the entire culture may appear involves selecting a treatment. Especially as more
more laden with one or another overdetermined per- specific therapies have become available, at times
sonality type. But as the cultural scales drop from the expensive or risky or difficult to apply, the need for
voyager’s eyes, the similarities in personality types accurate diagnosis-making has increased.
across cultures are remarkable. These differences and Diagnosis-making has come to have greater scien-
similarities are well known to students of cross-cultur- tific importance, especially for psychiatry. When little
al testing (93, 94). is known about the prognosis or treatment of a
Cross-cultural personality assessment in the clinical symptom or syndrome, diagnostic categorization be-
context is another matter, however. Even intracultural comes a serious investigative method. To be sure, it is
personality assessment in clinical settings is difficult. In at the most primitive level of science, a first-phase
the midst of crisis or acute psychiatric symptoms, endeavor to be followed later by comparative studies,
manifestations of personality disorder often emerge treatment outcome studies, epidemiology, research
only to disappear again as the distress recedes. Add to into etiology, and so forth. Primitive though it is,
this such factors as the foreign patient in a strange however, diagnostic classification is a key step in
cultural milieu, unfamiliar with psychiatric proce- pursuing epidemiological distribution, possible etio-
dures, or the clinician unfamiliar with a minority logical factors, and description of pathological proc-
group, and it is easy to perceive the difficulties in- ess, with the goal of being able ultimately to set
volved. accurate prognosis and select effective treatment.
My own practice is to remain desultory in append- Cross-cultural diagnosis has not presented great diffi-
ing a personality label across cultural boundaries in culties for biopathological processes (e.g., rheumatic
clinical settings. This is not an adequate solution to fever, cervical cancer, diphtheria). However, cross-
this difficult problem, however. One hopes that the cultural diagnosis of psychopathological processes has
increasing cross-cultural research into personality as- created dilemmas for psychiatry.
sessment will contribute to our knowledge in this area. Only recently has cross-cultural psychiatric diagno-
sis begun to be practical for clinical service and
Changes Over Time research. In the early days of this century, students of
psychiatric diagnosis despaired even about intracul-
Unlike the chemistry of neurotransmitters, the phar- tural diagnosis. Writing in 1904, Kraepelin observed
macology of psychoactive drugs, and the anatomy of that “our clinical concepts vary so widely that for the
the limbic system, cultures can and sometimes do forseeable future such comparison is possible only if
change rapidly. The notion of an entirely uniform, the observations are made by one and the same
traditional, and unchanging culture can exist only in observer” (96). It was another half century before
the mind of the observer; it does not exist in the real Galdston, in 1957, was able to write about the exis-
world. This constant feature of cultures creates mis- tence of “a national psychiatry,” yet he still empha-
chief for students of psychopathology across cultures. sized that there was not yet an international psychiatry
Further problems relate to the fact that, on the basis in the same sense that there was an international
of reports over the last century, psychopathology also surgery (97). As outlined in this paper, national diag-
changes. As Prince (95) has pointed out, the cause of nostic preferences still continue. Greater consensus in
this change is not obvious. Is it due to modifications in international diagnostic practice is underway. The
our diagnostic fashions? Or to several factors, such as World Health Organization has greatly facilitated this
increased education or decreased religiosity? Or to process.
cultural factors? Or is it due to the changing functions Why should we be concerned about the reliability of
of psychiatry in society (i.e., less involvement with psychiatric diagnosis across national and cultural
social control and more involvement with treatment)? boundaries? One reason is that clinicians are called on
Or do we now diagnose more of certain disorders, more often to diagnose and treat people from other
such as depression, because there are more beneficial cultures. Tourism, education abroad, economic migra-
things that we can do for these conditions? Perhaps tion, and refugee movements have greatly increased in
many or all of these elements may contribute to the the last few decades. Return to the home country for
observed changes. psychiatric care may not be desirable or even feasible.
Psychiatrists today must be prepared for this eventual-
ity.
DISCUSSION A second reason concerns more the science than the
art of psychiatry. Surawicz and Sandifer (37) described
Clinical diagnosis has long had two fundamental this problem as follows: “A report on the effectiveness
purposes for the art of medicine, both of them practi- of drug A on psychosis B in country C will be more

Am J Psychiatry 142:7, July 1985 803

PSYCHIATRIC DIAGNOSIS ACROSS CULTURAL BOUNDARIES

meaningful to a psychiatrist if psychosis B is identical in the short-term prognosis of schizophrenic psychoses. Schi-
with B in his country.” Of course the argument is zophr Bull 4:102-113, 1978
25. Lesse 5: Behavioral problems masking depression-cultural and
applicable not only to psychopharmacology but also to
clinical survey. Am J Psychother 33:41-53, 1979
etiology, epidemiology, prevention, and other forms of 26. Shore JH, Kinzie JD, Hampson JL, et al: Psychiatric epidemiol-
treatment. Much psychiatric research is difficult, time ogy of an Indian village. Psychiatry 36:70-81, 1973
consuming, and expensive. No country can afford to 27. Pattison EM, Kaufman E: Encyclopedic Handbook of Alcohol-
ignore psychiatric knowledge gleaned elsewhere. ism. New York, Gardner Press, 1982
28. Simons RC: The resolution of the Latah paradox. J Nerv Ment
Currently our training programs and much of our Dis 168:19S-206, 1980
clinical practice lag far behind the cross-cultural re- 29. Takano R: Anthropophobia and Japanese performance. Psychi-
search findings and demonstrated diagnostic tech- atry 40:259-269, 1977
niques. Over the remaining years of this century, the 30. Kline NS: Psychiatry in Indonesia. Am J Psychiatry 119:809-
815, 1963
cross-cultural tasks facing psychiatrists will be chal-
31. Levan I, Bilu Y: A transcultural view of Israeli psychiatry.
lenging ones. Their solution will contribute not only to Transcultural Psychiatr Research Rev 17:7-36, 1980
care of patients across cultural boundaries but to 32. Dennis PA: Grisi Siknis among the Miskito. Medical Anthropol-
theoretical issues in the field of psychiatry at large. ogy 5:445-SOS, 1982
33. Westermeyer J: Amok, in Extraordinary Disorders of Human
Behavior. Edited by Friedman CTH, Faguet R. New York,
REFERENCES Plenum, 1982
34. Weidman HH: Falling-out: a diagnostic and treatment problem
1. Collomb M: Bouff#{233}esd#{233}lirantesen psychiatric Africaine. Psy- viewed from a transcultural perspective. Soc Sci Med 13B:95-
chopathologie Africaine 1:167-239, 1965 112, 1979
2. Kroll J: Philosophical foundations of French and US nosology. 35. WestermeyerJ: On the epidemicity of amok. Arch Gem Psychia-
AmJ Psychiatry 136:1135-1138, 1979 try 28:873-876, 1973
3. Lachman JH: Swedish and American psychiatry: a comparative 36. Kenny MG: Paradox lost: the latah problem revisited. J Nerv
view. Am J Psychiatry 126:1777-1781, 1970 Ment Dis 171:159-167, 1983
4. Reich W: The spectrum concept of schizophrenia: problems for 37. Surawicz FG, Sandifer MG: Cross cultural diagnosis: a study of
diagnostic practice. Arch Gen Psychiatry 32:489-498, 1975 psychiatric diagnosis, comparing Switzerland, the United States,
S. Bellak L: Intercultural studies in search of a disease. Schizophr and the United Kingdom. mt J Soc Psychiatry 16:232-236,
Bull 1:6-9, 1975 1970
6. Kendell RE: Psychiatric diagnosis in Britain and the United 38. Sartorius N, Jablensky A, Shapiro R: Preliminary communica-
States. Br J Psychiatry Special Publication 9:453-461, 1975 tion: two-year follow-up of the patients included in the WHO
7. Stephens JH: Long-term course and prognosis in schizophrenia. International Pilot Study ofSchizophrenia. Psychol Med 7:529-
Semin Psychiatry 2:464-485, 1970 541, 1977
8. Stephens JH, Shaffer JW, Carpenter WT: Reactive psychoses. J 39. Sartorius N, Jablensky A, Gulbinat W, et al: Preliminary
Nerv Ment Dis 170:657-663, 1982 communication, WHO collaborative study: assessment of de-
9. Murase T, Johnson F: Naikan, Morita, and Western psychother- pressive disorders. Psychol Med 10:743-749, 1980
apy. Arch Gen Psychiatry 31:121-128, 1974 40. Fava GA, Kellner R, Munari F, et al: The Hamilton Depression
10. Xia Z, Zhang M: History and present status of modern Rating Scale in normals and depressives. Acta Psychiatr Scand
psychiatry in China. Chin Med J [Engi] 94:277-282, 1981 66:16-32, 1982
1 1. Chatel J, Joe B: Psychiatry in Spain: past and present. Am J 41. Kinzie JD, Manson SM, Vinh DT, et al: Development and
Psychiatry 132:1182-1186, 1975 validation of a Vietnamese-language depression rating scale.
12. Erinosho OA: Mental health delivery-systems and post-treat- AmJ Psychiatry 139:1276-1281, 1982
ment performance in Nigeria. Acta Psychiatr Scand 55:1-9, 42. Ebigbo P0: Development of a culture specific (Nigeria) screen-
1977 ing scale of somatic complaints indicating psychiatric distur-
13. Lippman D: Psychiatry in Ethiopia. Can Psychiatr Assoc J bance. Cult Med Psychiatry 6:29-43, 1982
21:383-388, 1976 43. Marsella AJ: Thoughts on cross-cultural studies on the epidemi-
14. Lin K, Kleinman A: Recent development of psychiatric epidemi- ology of depression. Cult Med Psychiatry 2:343-357, 1978
ology in China. Cult Med Psychiatry 5:135-143, 1981 44. Abad V, Boyce E: Issues in psychiatric evaluations of Puerto
15. Sethi BB, Gupta SC, Lal N: The theory and practice of Ricans: a socio-cultural perspective. J Operational Psychiatry
psychiatry in India. Am J Psychother 31:43-65, 1977 10:28-39, 1979
16. Dunham HW: Sociocultural studies of schizophrenia. Arch Gen 45. LeffJP: The cross-cultural study of emotions. Cult Med Psychi-
Psychiatry 24:206-214, 1971 atry 1:317-350, 1977
17. Redlich F, Kellert SR: Trends in American mental health. Am J 46. Gottheil E, Thornton CC, Exline RV: Appropriate and back-
Psychiatry 135:22-28, 1978 ground affect in facial displays of emotion: comparison of
18. Lipkowitz MH, Idupuganti 5: Diagnosing schizophrenia in normal and schizophrenic males. Arch Gem Psychiatry 33:565-
1980: a survey of US psychiatrists. Am J Psychiatry 140:52-55, 568, 1976
1983 47. Foulks EF: Interpretations of human affect. J Operational
19. Levine D: A cross-national study of attitudes toward mental Psychiatry 10:20-27, 1979
illness. J Abnorm Psychol 80:1 1 1-1 14, 1972 48. Ekman P, Friesen WV: Constancy across cultures in the face and
20. Westermeyer J, Doheny 5, Stone B: An assessment of clinical emotion. J Pers Soc Psychol 17:124-129, 1971
care for the alcoholic patient. Alcoholism: Clinical and Experi- 49. Gokulanathan KS, Vershese KP: Socio-cultural malnutrition. J
mental Research 2:53-57, 1978 Trop Pediatr 15:118-124, 1969
21. Warnes H: Cultural factors in Irish psychiatry. Psychiatr J Univ so. Coil CG, Spekoski C, Lester BM: Cultural and biomedical
Ottawa 4:329-335, 1979 correlates of neonatal behavior. Dcv Psychobiol 14:147-154,
22. Jablensky A, Sartorious N: Culture and schizophrenia. Psychol 1981
Med 5:113-124, 1975 51. Solomons HC: Standardization of the Denver developmental
23. Nichter M: Idioms of distress, alternatives in the expression of screening test on infants from Yucatan, Mexico. Int J Rehabii
psychosocial distress: a case study from South India. Cult Med Res 5:179-189, 1982
Psychiatry 5:379-408, 1981 52. Waidron I, Nowotarski M, Freimer M, et ai: Cross-cultural
24. Sartorius N, Jablensky A, Shapiro R: Cross-cultural differences variation in blood pressure: a quantitative analysis of the

804 Am J Psychiatry I 42: 7, July 1985

 JOSEPH WESTERMEYER

relationships of blood pressure to cultural characteristics, salt study of diagnosis of the mental disorders: a comparative
consumption and body weight. Soc Sci Med 16:419-430, 1982 psychometric assessment of elderly patients admitted to mental
53. Orley J, Wing JK: Psychiatric disorders in two African villages. hospitals serving Queens County, New York, and the former
Arch Gem Psychiatry 36:513-520, 1979 borough of Camberwell, London. Br J Psychiatry 126:560-
54. Marks IM: Research in neurosis, a selective review, 1 : causes 570, 1975
and courses. Psychol Med 3:436-454, 1973 76. Frias CA: A transcultural survey of psychiatric opinion on
ss. Stoker DH, Meadow A: Cultural differences in child guidance schizophrenia. Compr Psychiatry 15:225-231, 1974
clinic patients. Int J Soc Psychiatry 20:186-202, 1974 77. Shields J, Gottesman II: Cross-national diagnosis of schizophre-
56. Masserman JM, Carmichael HT: Diagnosis and prognosis in nia in twins. Arch Gem Psychiatry 27:725-730, 1972
psychiatry: with a follow-up study of the results of short-term 78. Giel R, d’Arrigo Busnello E, Climent CE, et al: The classifica-
general hospital therapy in psychiatric cases. J Ment Sci tion of psychiatric disorder: a reliability study in the WHO
84:893-946, 1938 collaborative study on strategies for extending mental health
57. Ash P: The reliability of psychiatric diagnosis. J Abnorm Soc care. Acta Psychiatr Scand 63:61-74, 1981
Psychol 44:272-275, 1949 79. Westermeyer J, Vang TF, Neider J: A comparison of refugees
58. Hunt WA, Wittson CL, Hunt EBA: A theoretical and practical using and not using a psychiatric service: an analysis of DSM-
analysis of the diagnostic process, in Current Problems in III criteria and self-rating scales in cross-cultural context. J
Psychiatric Diagnosis. Edited by Hock PN, Zubin J. New York, Operational Psychiatry 14:36-41, 1983
Grune & Stratton, 1953 80. Westermeyer J, Peake E: A ten-year follow up of alcoholic
59. Meehi PE, Rosen A: Antecedent probability and the efficiency Native Americans in Minnesota. Am J Psychiatry 140:189-
of psychometric signs, patterns, or cutting scores. Psychol Bull 194, 1983
52:194-216, 1955 81. Sethi BB, Prakash R: Cultural factors in the organization of
60. Mehlman B: The reliability of psychiatric diagnosis. J Abnorm multicentre trials. Mod Probl Pharmacopsychiatry 16:21-36,
Soc Psychol 47:557-560, 1952 1981
61. Cohen J: A coefficient of agreement for nominal scales. Educa- 82. Okuma T: Differential sensitivity to the effects of psychotropic
tional and Psychological Measurement 20:37-46, 1960 drugs: psychotics vs normals, Asians vs Western populations.
62. Fleiss JL: Measuring nominal scale agreement among many Folia Psychiatr Neurol Jpn 35:79-87, 1981
raters. Psychol Bull 76:378-382, 1971 83. Gould Si: Morton’s ranking of races by cranial capacity.
63. Fleiss JL: Statistical Methods for Rates and Ikpulations. New Science 200:503-509, 1978
York, John Wiley & Sons, 1973 84. Nissom MP, Schmidt KE: Land-Dyak concept of mental illness.
64. Spitzer RL, Cohen J, Fleiss JL, et al: Quantification of agree- MedJ Malaya 21:352-357, 1967
ment in psychiatric diagnosis. Arch Gen Psychiatry 17:83-87, 85. Fabrega H, Metzger D: Psychiatric illness in a small Ladino
1967 community. Psychiatry 31:339-351, 1968
65. Spitzer RL, Fleiss JL: A re-analysis of the reliability of psychiat- 86. Schmidt KE: Some concepts of mental illness of the Murut of
nc diagnosis. Br J Psychiatry 125:341-347, 1974 Sarawak. Int J Soc Psychiatry 14:24-31, 1968
66. Helzer JE, Clayton PJ, Parnbakian R, et al: Reliability of 87. Resner G, Hartog J: Concepts and terminology of mental
psychiatric diagnosis, II: the test-retest reliability of diagnosis disorder among Malays. J Cross Cultural Psychol I :369-38 1,
classification. Arch Gen Psychiatry 34:136-141, 1977 1970
67. Hudgens RW: The use of the term “Undiagnosed Psychiatric 88. Tseng WS: The development of psychiatric concepts in tradi-
Disorder.” Br J Psychiatry I 19:529-532, 1971 tional Chinese medicine. Arch Gem Psychiatry 29:569-575,
68. WeIner A, Liss JL, Robins E: Undiagnosed psychiatric patients, 1973
part III: the undiagnosable patient Br J Psychiatry 123:91-98, 89. Westermeyer J: Folk concepts of mental disorder among the
1973 Lao: continuities with similar concepts in other cultures and in
69. Tsuang MT: Schizophrenia around the world. Compr Psychia- psychiatry. Cult Med Psychiatry 3:301-317, 1979
try 17:477-481, 1976 90. Westermeyer J, Zimmerman R: Lao folk diagnoses for mental
70. Westermeyer J, Sines L: Reliability of cross-cultural psychiatric disorder: comparison with psychiatric diagnosis and assessment
diagnosis with an assessment of two rating contexts. J Psychiatr with psychiatric rating scales. Medical Anthropology 5:425-
Res 15:199-213, 1979 443, 1981
71. Copeland JRM, Kelleher MJ, Kellert JM, et al: Diagnostic 91. Rohrl VJ: Culture, cognition, and intellect: towards a cross-
differences in psychogeriatric patients in London and New cultural view of intelligence. J Psychol Anthropology 2:337-
York: United Kingdom-United States diagnostic project. Can 364, 1979
Psychiatr Assoc J 19:267-271, 1974 92. Hunt E: On the nature of intelligence. Science 219:141-146,
72. Leff JP: Transcultural influences on psychiatrists’ rating of 1983
verbally expressed emotion. Br J Psychiatry 125:336-340, 93. Diaz-Guerrero R: Socio-cultural premises, attitudes and cross-
1974 cultural research. mt J Psychol 2:79-87, 1967
73. Cooper JE, Kendell RE, Gurland BJ, et al: Cross-national study 94. Holtzman WH: Concepts and methods in the cross-cultural
of diagnosis of the mental disorders: some results from the first study ofpersonality development. Hum Dcv 22:281-295, 1979
comparative investigation. Am J Psychiatry 125(April 95. Prince R: The changing picture of depressive syndromes in
Suppi):21-29, 1969 Africa. Can J African Studies 1:177-192, 1968
74. Kendell RE, Cooper JE, Gourlay AJ, et al: Diagnostic criteria of 96. Kraepelin E: Lectures on Clinical Psychiatry. Edited by John-
American and British psychiatrists. Arch Gen Psychiatry stone T. New York, William Wood, 1904
25:122-130, 1971 97. Galdston I: International psychiatry. Am J Psychiatry 114:103-
75. Cowan DW, Copeland JR, Kelleher MJ, et al: Cross-national 108, 1957

Am J Psychiatry 1 42: 7, July 1985 805