Transatlantic variation in the attributed etiology of psychosis

Abstract

Background:

Differences in transatlantic perception of psychosis have been reported in the historical psychiatric literature.

Aims:

This study aims to determine if articles in the American Journal of Psychiatry (AJP) are more likely to attribute biological factors to the etiology of psychosis than those of the British Journal of Psychiatry (BJP).

Methods:

A systematic MEDLINE search for articles in the AJP and BJP from 2005 to 2007 identified 360 abstracts with psychosis and etiology-related words. Chi-square analyses were used to test differences in the proportion of attributed biological or psychosocial etiology of psychosis in each journal.

Results:

A greater proportion of abstracts (83/87) in the AJP attributed biological etiology of psychosis (χ² = 12.33, df = 1, p < 0.001), while a greater proportion in the BJP (16/44 abstracts) attributed psychosocial etiology (χ² = 19.76, df = 1, p < 0.001).

Conclusions:

The AJP tends to publish biomedical explanations of psychosis, while the BJP shows a relative preference for psychosocial theories.

Keywords

Psychosis etiology culture psychiatry

Introduction

Over the course of psychiatric history, competing explanations of mental illness have faded in and out of fashion according to time and place (Porter, 2002; Shorter, 1997). Consequently, it should not be surprising that differences in the transatlantic psychiatric perception of psychosis have been reported in the world psychiatric literature. A few examples illustrate this point. First, in the landmark US–UK Diagnostic Project (Cooper et al., 1972), psychiatric diagnosis in New York was found lacking when compared to psychiatric diagnosis in London. The authors called into question the ability of American psychiatrists to accurately diagnose schizophrenia, a point that was driven home in dramatic fashion by the embarrassing conclusion that ‘The New York concept of schizophrenia is not a useful one and is likely to inhibit fruitful research if it is widely adopted’ (Cooper et al., 1972, p. 125). Murray (1979) affirmed this statement when he wrote, ‘No area of psychiatric practice has attracted more criticism from overseas than American diagnostic habits’ (p. 256). A second example has to do with how rates of psychosis in minority populations have been perceived differently in the United States and United Kingdom. In the United States, reported rates of psychosis in African Americans have changed markedly over the last 200 years. Prior to the Civil War, mental illness in slaves was rarely reported (McCandless, 1996). Following emancipation, rates of insanity (i.e. psychosis) in the former slaves were reported to have increased dramatically (O’Malley, 1914) to the point that by the mid-20th century, African Americans were widely accepted to have higher rates of psychosis, particularly schizophrenia, than Whites (Malzberg, 1944). With the end of segregation in 1970, high rates of schizophrenia and psychosis were no longer reported in American Blacks, replaced by the perception that misdiagnosis and clinician error were presumed to account for differential rates of psychosis in minority ethnic groups (American Psychiatric Association, 2004). These changes in the reported rates of psychotic disorder in African Americans were not consistent over time, but changed according to social, cultural and historical imperatives (Jarvis, 2008). The American perspective was in marked contrast to recent studies from the United Kingdom and Western Europe where immigrants from the Caribbean and Africa (and elsewhere) were reported to have truly elevated rates of psychosis and schizophrenia independent of misdiagnosis and clinician error (Jarvis, 2007). A third example has to do with changes in the preferred explanation of psychosis in the United Kingdom compared to the United States. In the mid-20th century, biological explanations held sway in the United Kingdom and the United States (Moncrieff & Crawford, 2001). A recent study suggested that the British psychiatric literature in the last few decades has been leaning toward social and psychological explanations of psychosis (Leff, 2008) compared to the American literature in which there has been a more consistent tradition of research into the biologic underpinnings of mental illness and a paucity of attention to social causes (Brodie & Sabshin, 1973; Pincus, Henderson, Blackwood, & Dial, 1993). The preoccupation with biological etiology in the United States was exemplified by the ‘Decade of the Brain’ in the 1990s (Goldstein, 1990), with its attendant focus by American psychiatry on neuroscience and related fields to the exclusion of psychosocial concepts (Jones & Mendell, 1999). The current presidential focus, the Brain Research through Advancing Innovative Neurotechnologies (BRAIN) initiative, may be yet another manifestation of the same trend (National Institutes of Health, 2013). The emphasis on comparing American and British psychiatry is warranted given the dominance of these countries in the mental health fields: the United States and the United Kingdom, with approximately 5% of the world’s population, contributed more than 50% of the world’s mental health-related publications available in the ISI Web of Science database from 1992 to 2001 (Saxena, Paraje, Sharan, Karam, & Sadana, 2006). Canada, while a much smaller player, is of interest as a third country that is a geographic neighbor to the United States, but shares cultural ties to the United Kingdom as part of the Commonwealth. Given the history of transatlantic differences in psychiatry, this study examined variations in the orientation toward the attributed etiology of psychosis between the United States, the United Kingdom, and Canada. To meet this aim, we systematically evaluated the content of the abstracts and titles of published articles in three national psychiatric association journals. Specifically, this study tested the hypothesis that a greater proportion of articles in the American Journal of Psychiatry (AJP) attributes biological factors to the etiology of psychosis, while the British Journal of Psychiatry (BJP) attributes etiology more often to social factors, and the Canadian Journal of Psychiatry (CJP) falls somewhere in between.

Method

Step 1

A systematic MEDLINE search of the AJP, BJP and CJP from 2005 to 2007 identified articles with psychosis and etiology-related keywords in the abstracts or titles. Key words for psychosis included: psychosis, psychoses, psychotic, psychotic disorders, schizophrenia, schizophrenic, schizoaffective, insanity, madness, dementia praecox, schizophreniform, schizotypal, schizotypy and serious mental illness. Key words and stems for etiology included: Etiolog –, aetiolog –, rate, rates, rationale, marker –, predict –, relat –, associat –, correlat –, covar –, path –, link –, caus –, origin –, vulner –, determin –, mechanism –, risk – and due to. Key words were derived from pilot data obtained by identifying and test rating relevant MEDLINE abstracts from the AJP, BJP and CJP from 1994, with additional key words added by consensus based on the authors’ knowledge of the literature.

Step 2

Raters, blind to journal of origin, determined whether in fact causality was implied upon reading the abstracts and titles. By implied, we judged that at an epistemological level, and within the cultural context of psychiatry, reported correlations between biological or social factors and psychosis would be likely to create a causal link in the minds of readers, editors, reviewers and researchers alike. Raters categorized the abstracts in which causality was implied as offering either a biological or non-biological (social or other) etiology of psychosis. As for the key words in Step 1, definitions of biological and social etiologies of psychosis were derived from pilot data obtained by identifying and test rating relevant MEDLINE abstracts from the AJP, BJP and CJP from 1994. Biological etiology of psychosis meant that causality was implied for at least one of the following terms in the abstracts or titles: biological, physical, universal, genetics, genetic, genes, genome, DNA, heredity, innate, inborn, family history, brain, brain development, brain asymmetry, brain structure, brain morphology, central nervous system (CNS), CNS development, perinatal or obstetrical complications, diet or nutrition, cell migration, imaging, neurotransmitters, dopamine, dopaminergic system, receptors, cognitive development, neuropsychological or neuropsychiatric. Similarly, social etiology of psychosis meant that causality was implied for at least one of the following terms in the abstracts or titles: social, social capital, social adversity, social defeat, social deprivation, social disadvantage, socioeconomic status (SES), urbanicity, ethnic density, poverty, racism, discrimination, prejudice, bigotry, stigma, migrant, migrants, migration, immigrant, immigrants, immigration, emigrant, emigrants, emigration, refugee, refugees, asylum seekers, ethnicity, ethnic, ethno-racial, race, skin color, Black, Black Briton, Black American, Black Canadian, African, African American, Afro-Canadian, Afro-Caribbean, Caribbean, West Indian, minority, religion or religiosity. Words for persons of African descent were included in the search terms due to studies from the United States and the United Kingdom that have focused on psychosis in members of this minority. Any causal terms that were neither biological or social (as defined above) were put into a third category designated ‘Other Etiology’. Many of these ‘Other’ causes of psychosis implied a cognitive or developmental etiology (i.e. psychological origin), so Social and Other etiologies were combined in this study to permit larger numbers for statistical comparison. The combined social and other category was called ‘psychosocial etiology’ and will be referred to as such throughout this article. For some papers in which more than one etiologic factor was mentioned, raters made a judgment call and classified the papers according to what they interpreted to be the predominant attributed etiology. As a result, biological and psychosocial categories were mutually exclusive, that is, any given paper could only be assigned to one category. Difficult abstracts, with multiple or ambiguous attributed etiology of psychosis, were assigned by consensus after raters consulted with the first author (G.E.J.). In practice, the number of these difficult abstracts was small (less than 5); most papers were assigned by the raters and with reasonable reliability as outlined in the next section.

Step 3

Inter-rater reliability was determined for select key study variables to ensure that raters were coming to similar conclusions regarding (1) whether or not psychosis and etiology-related concepts were present in abstracts or titles, (2) whether or not etiology of psychosis was implied in the abstracts and (3) whether the implied etiology was biological, social or other. Data for 50 abstracts were randomly selected for reliability testing. Kappas were acceptable at k = 1.0 for psychosis mentioned in abstract or title, k = 0.96 for etiology mentioned in abstract or title, k = 0.88 for causality implied in the abstract or title and k = 0.73 for assigned biological etiology of psychosis.

Step 4

Using IBM SPSS Statistics Version 20, chi-square analyses tested the significance of differences in the proportion of attributed biological or psychosocial etiology of psychosis in each journal.

Summary

We recognized that our assignment of etiology, either biological or psychosocial, to abstracts identified by keywords would be difficult to replicate. For this reason, and as mentioned above, the investigators practiced their assignments of etiology on abstracts from 1994 prior to completing the data collection for this study. They compared their results with acceptable kappa values. Papers with multiple or ambiguous etiology were assigned by consensus.

Results

From 2005 to 2007, 360 articles were identified with psychosis and etiology-related keywords in the abstracts or titles of which 191 were from the AJP, 126 from the BJP and 43 from the CJP. Of these, 149 were judged to attribute a causal link of some kind to psychosis (see Table 1). Table 2 compares these findings statistically: 83/87 (95.4%) abstracts attributed a biological etiology to psychosis in the AJP compared to 36/48 (75%) in the BJP (χ² = 12.33, df = 1, p < 0.001), and 5/85 (5.9%) abstracts attributed a psychosocial etiology to psychosis in the AJP compared to 16/44 (36.4%) in the BJP (χ² = 19.76, df = 1, p < 0.001). A review of the actual abstracts showed that of the nine with attributed social etiology in the BJP (see Table 1), four discussed psychosis in immigrants or migration and three explicitly discussed the social causes of psychosis. By contrast, only one abstract in the AJP mentioned these topics, and the authors were from Europe. In fact, of the five articles from the AJP with attributed psychosocial etiology, three of the lead authors were from Europe indicating almost no endorsement of psychosocial etiology of psychosis among American contributors to the AJP during the time period of this study. Discrepancies in the total number of abstracts for different models were the result of missing values that deleted a small number of abstracts from some analyses. The small number of CJP abstracts (n = 8 to 12 depending on the analysis) precluded separate comparison, but when Canadian abstracts were merged with American or British abstracts, the overall outcome was unchanged; however, the pattern of Canadian results was similar to the British (9/12, or 75%, for biological etiology and 3/11, or 27.3%, for psychosocial etiology).

Table 1.

Attributed etiology of psychosis by journal (N = 149 abstracts with an implied causal link to psychosis of any kind)^a.

	AJP	BJP	CJP
Attributed biological etiology (n = 147)	83/87 (95.4%)	36/48 (75.0%)	9/12 (75.0%)
Attributed social etiology (n = 139)	3/85 (3.5%)	9/43 (20.9%)	3/11 (27.3%)
Attributed other etiology (n = 139)	2/84 (2.4%)	9/44 (20.5%)	0/11 (0.0%)
Combined social and other etiologies (n = 140)	5/85 (5.9%)	16/44 (36.4%)	3/8 (27.3%)

AJP: American Journal of Psychiatry; BJP: British Journal of Psychiatry; CJP: Canadian Journal of Psychiatry.

Table 2.

Chi-squared statistical tests: attributed etiology by journal^a,b.

	AJP	BJP	Pearson chi-square	df	p
Biological etiology (n = 135)	83/87	36/48	12.3	1	<.0001
Psychosocial etiology (n = 129)	5/85	16/44	19.76	1	<.0001

AJP: American Journal of Psychiatry; BJP: British Journal of Psychiatry; CJP: Canadian Journal of Psychiatry.

All data were not available for all abstracts.

Abstracts from the CJP were not included in this analysis because of small numbers.

Discussion

A key finding of this study is that all three national journals publish articles that predominantly report biological explanations of psychosis (i.e. between 75% and 95% of abstracts and titles). These findings suggest that the culture of contemporary psychiatry in the United States, the United Kingdom and Canada, as represented by these flagship journals, is overwhelmingly one of biological explanations and etiologies with respect to psychosis. A second key finding of this study is that this biological orientation was most pronounced in American psychiatry, as represented by abstracts published in the AJP, where the biological underpinnings of psychosis were present in virtually every relevant abstract from 2005 to 2007. A third key finding is that the BJP published a greater proportion of articles with a psychosocial explanation of psychosis, but this was only a matter of degree. In other words, British psychiatry may be more open to asserting psychosocial etiologies of psychosis, but these still represent a minority position. This finding is consistent with recent work that found differential publishing trends between the AJP and the BJP in the years 1951–2005, with a higher percentage of psychosocial articles in the BJP than the AJP since about the late 1980s (Leff, 2008).

These results imply that attempts by psychiatry to adopt biopsychosocial models of the etiology of psychosis have met with limited success such that biological models have persisted to a large degree, especially in the United States but also in Canada and the United Kingdom. Why these trends? The culture of psychiatry in the United Kingdom may be somewhat different from that of the United States: in the United Kingdom, the psychiatric literature has been reporting for several decades that elevated rates of psychosis in immigrants are real rather than due to artifacts of misdiagnosis, and that this is due to social rather than genetic factors (Kirkbride et al., 2008; Kirkbride et al., 2006). In contrast, in the United States, the trend has been to identify key biogenetic factors, presumed to be universal across ethnic groups, that will not single out one group over another as being more prone to developing psychotic disorders (Jarvis, 2008). These different contexts may be fostering different research agendas in the two countries. In the United Kingdom, a social phenomenon, such as immigration, has influenced the psychiatric literature and changed the way psychiatrists have come to understand the etiology and development of presumed biological illnesses like psychosis and schizophrenia. On the other hand, in the United States, attributed etiology of psychosis has veered away from potentially controversial social theories of causation related to race, immigration and poverty and toward a less socially embedded model of biogenetics in which macrosocial influence on illness onset may be conveniently neglected.

Furthermore, American psychiatry may be driven more by vested interests that have a stake in keeping biological explanations high on the agenda, whereas in the United Kingdom, there has been greater regulation of such interests. For example, some scholars have identified the US pharmaceutical industry as being a particularly powerful player in maintaining the psychiatric research focus on biological factors (Whitaker, 2010).

These national variations in the psychiatric literature may simply reflect a difference in the dominant narratives between the United States and the United Kingdom regarding the determinants of negative individual outcomes. It has been argued that the United Kingdom has a stronger tradition of explaining factors such as poverty, unemployment, crime and illness as the consequence of macro-level social inequalities, whereas in the United States such outcomes are routinely framed as the consequence of individual-level failings (Luhrman, 2007; Wilkinson & Pickett, 2010). This British tradition of explaining ill health in social terms is embodied in the seminal work of respected scholars such as Richard Wilkinson (Wilkinson & Pickett, 2010) and Sir Michael Marmot (Marmot et al., 1991), whose research indicates a strong relationship between social inequality and negative health outcomes. This British social tradition is also evidenced by political and societal consensus regarding the provision of universal health care through the National Health Service. No such consensus or parallel organization exists in the United States. In other words, the observed difference in focus between American and British studies may be part of wider national narratives that differentially focus upon macro-level social or individual-level factors as determinants of outcomes. All of these factors may consciously or unconsciously influence activities of funding agencies, research institutions, psychiatric researchers, as well as the preferences of journal editors, heads of departments and funding reviewers in the kinds of research and publications they encourage.

A number of limitations should be noted. First, do the conclusions based on data from three psychiatric journals apply more widely to the profession? We specifically chose the AJP, BJP and CJP because they are published by national psychiatric associations and, consequently, enjoy a wide distribution among the psychiatrists of the United States, United Kingdom and Canada. Due to their easy accessibility and presumed relative influence, the content of these journals seem likely to represent national trends, at least in a very general way. Second, we only examined abstracts and titles rather than the complete articles during the specified time period. However, our reasoning was that abstracts and titles include concise information that would more likely be read and internalized by the average psychiatrist, and thus may represent a condensation of knowledge trends in the field. Third, we only used MEDLINE searches and did not examine other databases (e.g. PsycINFO or PubMed) and hence may have missed some additional articles, but we reasoned that since the AJP, BJP and CJP are MEDLINE indexed, it is unlikely that other databases would have additional articles from these journals. In any event, any leakage of articles would be random rather than systematic. Fourth, we did not examine the articles’ reference sections, but limited our search to MEDLINE itself over the years of the study. Finally, the terms used in the literature search to represent etiology-related concepts may have overlapped with non-etiological correlational factors. For example, a neurological correlate of schizophrenia reported in an abstract or title is not necessarily evidence of causation, but we believed that at an epistemological level, and within the cultural context of psychiatry, it may imply causation or may be inferred to imply so in the minds of readers, editors, reviewers and researchers alike. Furthermore, to focus on biological or psychosocial correlates associated with psychosis indicates a belief, or at least a relative importance, given to that variable by the authors of the article. Why else would it be deemed worthy of inclusion in the space-limited abstract for all to read at a glance?

Future studies may examine additional journals over a longer period of time using a larger set of abstracts. Comparisons of perceptions across different eras (e.g. 1950s vs 1990s) and across different countries and languages may also prove to be informative, as would an assessment of papers with more than one implied etiology (such as bio-social origins of psychosis) in an effort to elaborate a more nuanced multifactorial approach to mental disorders in psychiatry. A survey of British and North American researchers, asking them to weigh the relative importance of biogenetic and psychosocial etiologic factors, would be a more direct way to uncover transatlantic variations in the attribution of etiology of psychosis.

By focusing on the difference between biological and psychosocial causes of psychosis in this article, we do not wish to unintentionally promote body–mind dualism. We recognize that biological and psychosocial factors are closely linked and have tremendous overlap (Ryder, Ban, & Chentsova-Dutton, 2011). Rather, we are trying to draw attention to the culture of psychiatry itself and how attributions of etiology, or cause, may be relatively more or less favored in one professional culture than another, even in those from the United States and the United Kingdom, which share similar histories, language, values and cultural symbols.

In summary, the findings support the hypotheses regarding the publication patterns of the American, British and Canadian Journals of Psychiatry. Based on abstracts and titles, all journals focused overwhelmingly on biological etiology of psychosis, meaning that general acceptance of the biopsychosocial model of explanation of psychosis has been pre-empted by a bio–bio–bio model. That said, significantly more articles in the BJP highlighted social etiologies, signifying that the biopsychosocial model may carry more weight in the United Kingdom. These findings make plausible the assertion that American psychiatry remains a culture of biological inquiry into the causes of psychosis, whereas recent waves of immigration to the United Kingdom have provoked a re-evaluation of psychosocial causes of psychosis. Importantly, the overall conclusion of this article is that scientific inquiry is not immune from the social and historical context of the larger society. While biological etiologies dominate the mainstream psychiatric understanding of psychosis in the United States, United Kingdom and Canada, social paradigms may be on the upswing, at least in the United Kingdom (Priebe, Burns, & Craig, 2013).

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

American Psychiatric Association. (2004). Practice guideline for the treatment of patients with schizophrenia, second edition. The American Journal of Psychiatry, 161(Suppl. 2), 1–50.

Brodie

H. K. H.

Sabshin

(1973). An overview of trends in psychiatric research: 1963–1972. The American Journal of Psychiatry, 130, 1309–1318.

Cooper

J. E.

Kendell

R. E.

Gurland

B. J.

Sharpe

Copeland

J. R. M.

Simon

(1972). Psychiatric diagnosis in New York and London. Oxford, UK: Oxford University Press.

Goldstein

(1990). The decade of the brain. Neurology, 40, 321.

Jarvis

G. E.

(2007). The social causes of psychosis in North American psychiatry: A review of a disappearing literature. The Canadian Journal of Psychiatry, 52, 287–294.

Jarvis

G. E.

(2008). Changing psychiatric perception of African Americans with psychosis. The European Journal of American Culture, 27, 227–252.

Jones

E. G.

Mendell

L. M.

(1999). Assessing the decade of the brain. Science, 284, 739.

Kirkbride

J. B.

Barker

Cowden

Stamps

Yang

Jones

P. B.

Coid

J. W.

(2008). Psychoses, ethnicity and socio-economic status. British Journal of Psychiatry, 193, 18–24.

Kirkbride

J. B.

Fearon

Morgan

Dazzan

Morgan

Tarrant

… Jones

P. B.

(2006). Heterogeneity in incidence rates of schizophrenia and other psychotic syndromes: Findings from the 3-center AESOP study. Archives of General Psychiatry, 63, 250–258.

10.

Leff

(2008). Climate change in psychiatry: Periodic fluctuations or terminal trend? In Morgan

McKenzie

Fearon

(Eds.), Society and psychosis (pp. 11–22). Cambridge, UK: Cambridge University Press.

11.

Luhrman

T. M.

(2007). Social defeat and the culture of chronicity: Or, why schizophrenia does so well over there and so badly here. Culture, Medicine and Psychiatry, 31, 135–172.

12.

Malzberg

(1944). Mental disease among American negroes: A statistical analysis. In Klineberg

(Ed.), Characteristics of the American Negro (pp. 373–399). New York, NY: Harper & Brothers.

13.

Marmot

M. G.

Smith

G. D.

Stansfield

Patel

North

Head

… Feeney

(1991). Health inequalities among British civil servants: The Whitehall II study. The Lancet, 337, 1387–1393.

14.

McCandless

(1996). Moonlight, magnolias, & madness. Chapel Hill: The University of North Carolina Press.

15.

Moncrieff

Crawford

M. J.

(2001). British psychiatry in the 20th century – Observations from a psychiatric journal. Social Science & Medicine, 53, 349–356.

16.

Murray

R. M.

(1979). A reappraisal of American psychiatry. The Lancet, 313, 255–258.

17.

National Institutes of Health. (2013). Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative. Retrieved from http://www.nih.gov/science/brain/

18.

O’Malley

(1914). Psychoses in the colored race: A study in comparative psychiatry. American Journal of Insanity, 71, 309–337.

19.

Pincus

H. A.

Henderson

Blackwood

Dial

(1993). Trends in research in two general psychiatric journals in 1969–1990: Research on research. The American Journal of Psychiatry, 150, 135–142.

20.

Porter

(2002). Madness: A brief history. New York, NY: Oxford University Press.

21.

Priebe

Burns

Craig

T. K. J.

(2013). The future of academic psychiatry may be social. The British Journal of Psychiatry, 202, 319–320.

22.

Ryder

A. G.

Ban

L. M.

Chentsova-Dutton

Y. E.

(2011). Towards a cultural-clinical psychology. Social & Personality Psychology Compass, 5, 960–975.

23.

Saxena

Paraje

Sharan

Karam

Sadana

(2006). The 10/90 divide in mental health research: Trends over a 10-year period. British Journal of Psychiatry, 188, 81–82.

24.

Shorter

(1997). A history of psychiatry: From the era of the asylum to the age of Prozac. Toronto, Ontario, Canada: John Wiley.

25.

Whitaker

(2010). Anatomy of an epidemic. New York, NY: Broadway Paperbacks.

26.

Wilkinson

Pickett

(2010). The spirit level: Why greater equality makes societies stronger. New York, NY: Bloomsbury Press.