Schizophreniaforum - Print Discussion

Dana March		James Kirkbride		Wim Veling

On 17 March 2009, we hosted a live discussion of fresh new ideas in the epidemiology of schizophrenia. Discussion leaders Dana March of Columbia University, James Kirkbride of the University of Cambridge, and Wim Veling of Parnassia Psychiatric Institute delivered a wide-ranging discussion of social factors such as migration, ethnicity, and urbanicity, but also asked how this research could benefit from genetic insights. Finally, they discussed possible biological mechanisms that might transduce social factors into psychosis.

1. Elucidation of several factors well established: migration, ethnicity, urbanicity
The study of social causes of schizophrenia, once a dominant focus of attention, has been reinvigorated in the current era. This growing area of study draws on evidence from analyses of migration, ethnicity, and urbanicity. Recent and ongoing work grows out of interest in work linking migration and schizophrenia. Evidence indicates that migration itself is not the important exposure, since second-generation migrants appear more at risk (Cantor-Graae and Selten, 2005).

Studies of ethnicity and race, stimulated by the migration work, show that racial and ethnic minorities are at markedly increased risk of schizophrenia, compared to their majority native-born counterparts (Fearon et al., 2006; Veling et al., 2006; Bresnahan et al., 2007). Interestingly, ethnic minorities who live in areas with greater concentrations of other ethnic minorities are less at risk than ethnic minorities who live in areas with few ethnic minorities (Boydell et al., 2001; Veling et al., 2007). One possible explanation is the lack of discrimination, which has been associated with increased risk (Veling et al., 2007). Still, these findings leave many questions unanswered, and evidence is lacking across settings.

Finally, studies of urbanicity indicate that people born in urban areas are about two times more likely to develop schizophrenia than those born in less densely populated areas (e.g., Mortensen and Pedersen, 1999). However, what it is about being born in an urban area remains elusive—the relevant exposures may be physical (e.g., infection, toxins, nutrition) or social (e.g., crowding, adversity) in nature. Moreover, it is unclear whether this finding holds across a range of settings, since the bulk of evidence comes from Western Europe (March et al., 2008).

Together, studies of migration, ethnicity, and urbanicity frame a set of compelling questions that are being taken up by a new wave of research. Here we address the conceptual and methodological aspects of this work that currently make it one of the most interesting frontiers of schizophrenia research.

2. From risk indicators to causes
One of the central paradoxes in the social epidemiology of schizophrenia and other psychoses is that while urbanicity and black and minority ethnic status are associated with increased rates of disorder, they are ubiquitous “exposures” with little specificity. Many millions of people are exposed to these factors, but roughly only 50 per 100,000 people per year will go on to develop the disorder (Kirkbride et al., 2006). These factors may be better described as risk indicators—associated with an increased risk of psychoses, but not of themselves etiologically relevant. Rather, they lie on the causal pathway and present markers of risk for factors which lie closer to the causal genetic, individual, and environmental effects we will need to elicit before it will be possible to think about identifying environmental factors which offer potentially tangible public health prevention targets. Current research should focus on identifying risk factors for psychoses with greater specificity to allow us to resolve this paradox. So far, there have been important contributions from two streams of research, elucidating distinct genetic and environmental risks for psychotic disorders (McGrath et al., 2004; International Scizophrenia Consortium, 2008). To understand the likely complexity in the etiology of psychosis, we now need to move toward investigating how social factors interact with underlying genetic vulnerability to increase the risk of psychosis for some, but not all, individuals. Important contributions in this regard are beginning to emerge (Caspi et al., 2005), but we will need to develop a better understanding of the likely multifactorial, multilevel structure of exposures which increase and protect against the onset of psychosis across the life course, elucidating the critical time periods where these factors have the greatest effect (March and Susser, 2006).

3. Possible biological mechanisms
There is general consensus that heightened dopaminergic transmission plays a central role in schizophrenia. Several studies demonstrated that (untreated) schizophrenia patients have an increased mesolimbic dopamine release after acute amphetamine challenge (Laruelle et al., 1996; Abi-Dargham et al., 1998), suggesting an abnormal responsiveness of dopaminergic neurons. The question remains as to what causes this hypersensitivity to dopamine. It has been speculated that dopamine dysregulation arises from genetic factors, such as the valine-to-methionine (Val/Met) polymorphism in the catechol-O-methyl transferase (COMT) gene, which is involved in dopamine metabolism (Williams et al., 2007). Dopamine dysregulation might also be caused by early environmental insults such as prenatal infections, malnutrition, or obstetric complications, and by social stress in adolescence or early adult life (Howes et al., 2004). Several neurobiological models suggest links among stress, the hypothalamic-pituitary-adrenal (HPA) axis, and dopamine activity (Corcoran et al., 2003). Animal experiments have shown that exposure to stress as well as the biological induction of HPA activation enhance the behavioral response of rats to dopamine agonists (Covington and Miczek, 2001), and lead to mesolimbic dopaminergic hyperactivity (Tidey and Miczek, 1996). Thus, social stress may lead to (further) dysregulation of the dopamine system through an augmenting effect of the HPA axis on dopamine synthesis and release (Walker and Diforio, 1997). At the same time, it is likely that there is a reciprocal effect such that heightened sensitivity to dopamine influences HPA activation and thus renders the individual hyper-responsive to stress, consistent with recent human studies that found that individuals who are genetically vulnerable to psychosis are more emotionally and behaviorally sensitive to daily life stress than are healthy controls (Myin-Germeys et al., 2001). Consequently, if the results of the animal experiments can be extended to humans, it is possible that chronic exposure to social stress leads to disturbances in dopamine function and further to the development of psychosis.

1. Understanding the broader social processes which underpin social epidemiology.
2. Understanding historical timing of social events and historical changes in social factors.
3. Understanding critical timing of exposures across the life course.
4. Understanding interactions (interplay) between genes and environment to increased risk of schizophrenia.
5. Elucidating biological processes which link social factors to onset of schizophrenia—dopamine pathways, hippocampus changes, epigenetics, etc.
6. Understanding the contextual nature of risk factors—different factors in different populations (both genetic and environmental variation).
7. Construct refinement in gXe interactions—types of interactions.

We need a better understanding of the phenotype, genotype, and envirome (greater specificity). What to target first? Candidate genes? What putative mechanisms are worth pursuing?

How do we design studies to deal with the complexities in etiology, and the particular complexities in social etiology?

How do we reconcile different contextual mechanisms in different settings to have meaningful public health implications?

1. Construct refinement/meaning
What are we actually capturing when we measure urbanicity or ethnic density? Examining critically the constructs of interest is a key step in understanding mechanisms. Part of this is trying to understand the broader social processes that create the conditions we ultimately measure as exposures of interest (e.g., population density, percent minority composition). With urbanicity, for instance, how do areas become densely populated? What are the patterns of urbanization and suburbanization in a given geographic location?

2. Study designs
We propose that it is necessary to design studies that investigate multiple levels of causation, because the complex etiology of schizophrenia involves factors acting at the level of genes, individuals, ethnic/racial groups, and neighborhoods. It is likely that higher-order interactions between genes and environment, and/or genes and genes, and/or environment and environment, determine the development of schizophrenia. Many different pathways are conceivable. This complexity requires research with carefully specified hypotheses and large samples in varying social contexts. But how can we accomplish that?

3. Time
Both historical time and timing of exposure during the life course are critical to understanding contextual effects. Indeed, period and cohort effects may exist with contextual exposures; beyond that, to understand better broader social processes, a historical view is beneficial. For example, in the work being conducted in Oakland, urbanization and suburbanization during the 1940s and 1950s created particular and identifiable patterns of urbanicity by race (hence “ethnic” density) and SES. Different patterns may be observed in other locations at other times. Understanding particulars may just help us wrap our minds around the generalizable; they are not mutually exclusive.

With respect to timing of exposure, we know population density has the greatest effect at birth. Ethnic density seems to operate around the time of illness onset—or does it? We need to examine these contextual exposures systematically across the life course in order to understand potential mechanisms.

Construct refinement by understanding broader social processes with a historical perspective and consideration of the individual life course will also inform the potential confounds we measure and include in our models.

If further studies would find more (specific) evidence for a social etiology of schizophrenia, public health interventions might be developed to prevent psychosis. These interventions may include strategies to diminish social isolation, increase social capital, and facilitate access to this capital. It may be feasible to strengthen social structures within social groups and within neighborhoods.

References:
Cantor-Graae E, Selten JP. Schizophrenia and migration: a meta-analysis and review. Am J Psychiatry. 2005 Jan 1;162(1):12-24. Abstract

Kirkbride JB, Fearon P, Morgan C, Dazzan P, Morgan K, Tarrant J, Lloyd T, Holloway J, Hutchinson G, Leff JP, Mallett RM, Harrison GL, Murray RM, Jones PB. Heterogeneity in incidence rates of schizophrenia and other psychotic syndromes: findings from the 3-center AeSOP study. Arch Gen Psychiatry. 2006 Mar 1;63(3):250-8. Abstract

Veling W, Selten JP, Veen N, Laan W, Blom JD, Hoek HW. Incidence of schizophrenia among ethnic minorities in the Netherlands: a four-year first-contact study. Schizophr Res. 2006 Sep 1;86(1-3):189-93. Abstract

Bresnahan M, Begg MD, Brown A, Schaefer C, Sohler N, Insel B, Vella L, Susser E. Race and risk of schizophrenia in a US birth cohort: another example of health disparity? Int J Epidemiol. 2007 Aug 1;36(4):751-8. Abstract

Boydell J, van Os J, McKenzie K, Allardyce J, Goel R, McCreadie RG, Murray RM. Incidence of schizophrenia in ethnic minorities in London: ecological study into interactions with environment. BMJ. 2001 Dec 8;323(7325):1336-8. Abstract

Veling W, Susser E, van Os J, Mackenbach JP, Selten JP, Hoek HW. Ethnic density of neighborhoods and incidence of psychotic disorders among immigrants. Am J Psychiatry. 2008 Jan 1;165(1):66-73. Abstract

Veling W, Selten JP, Susser E, Laan W, Mackenbach JP, Hoek HW. Discrimination and the incidence of psychotic disorders among ethnic minorities in The Netherlands. Int J Epidemiol. 2007 Aug 1;36(4):761-8. Abstract

Mortensen PB, Pedersen CB, Westergaard T, Wohlfahrt J, Ewald H, Mors O, Andersen PK, Melbye M. Effects of family history and place and season of birth on the risk of schizophrenia. N Engl J Med. 1999 Feb 25;340(8):603-8. Abstract

March D, Hatch SL, Morgan C, Kirkbride JB, Bresnahan M, Fearon P, Susser E. Psychosis and place. Epidemiol Rev. 2008 Jan 1;30():84-100. Abstract

McGrath J, Saha S, Welham J, El Saadi O, MacCauley C, Chant D. A systematic review of the incidence of schizophrenia: the distribution of rates and the influence of sex, urbanicity, migrant status and methodology. BMC Med. 2004 Apr 28;2():13. Abstract

[No authors listed]. Rare chromosomal deletions and duplications increase risk of schizophrenia. Nature. 2008 Sep 11;455(7210):237-41. Abstract

Caspi A, Moffitt TE, Cannon M, McClay J, Murray R, Harrington H, Taylor A, Arseneault L, Williams B, Braithwaite A, Poulton R, Craig IW. Moderation of the effect of adolescent-onset cannabis use on adult psychosis by a functional polymorphism in the catechol-O-methyltransferase gene: longitudinal evidence of a gene X environment interaction. Biol Psychiatry. 2005 May 15;57(10):1117-27. Abstract

March D, Susser E. The eco- in eco-epidemiology. Int J Epidemiol. 2006 Dec 1;35(6):1379-83. Abstract

Laruelle M, Abi-Dargham A, van Dyck CH, Gil R, D'Souza CD, Erdos J, McCance E, Rosenblatt W, Fingado C, Zoghbi SS, Baldwin RM, Seibyl JP, Krystal JH, Charney DS, Innis RB. Single photon emission computerized tomography imaging of amphetamine-induced dopamine release in drug-free schizophrenic subjects. Proc Natl Acad Sci U S A. 1996 Aug 20;93(17):9235-40. Abstract

Abi-Dargham A, Gil R, Krystal J, Baldwin RM, Seibyl JP, Bowers M, van Dyck CH, Charney DS, Innis RB, Laruelle M. Increased striatal dopamine transmission in schizophrenia: confirmation in a second cohort. Am J Psychiatry. 1998 Jun 1;155(6):761-7. Abstract

Williams HJ, Owen MJ, O'Donovan MC. Is COMT a susceptibility gene for schizophrenia? Schizophr Bull. 2007 May 1;33(3):635-41. Abstract

Howes OD, McDonald C, Cannon M, Arseneault L, Boydell J, Murray RM. Pathways to schizophrenia: the impact of environmental factors. Int J Neuropsychopharmacol. 2004 Mar 1;7 Suppl 1():S7-S13. Abstract

Corcoran C, Walker E, Huot R, Mittal V, Tessner K, Kestler L, Malaspina D. The stress cascade and schizophrenia: etiology and onset. Schizophr Bull. 2003 Jan 1;29(4):671-92. Abstract

Covington HE, Miczek KA. Repeated social-defeat stress, cocaine or morphine. Effects on behavioral sensitization and intravenous cocaine self-administration "binges". Psychopharmacology (Berl). 2001 Dec 1;158(4):388-98. Abstract

Tidey JW, Miczek KA. Social defeat stress selectively alters mesocorticolimbic dopamine release: an in vivo microdialysis study. Brain Res. 1996 May 20;721(1-2):140-9. Abstract

Walker EF, Diforio D. Schizophrenia: a neural diathesis-stress model. Psychol Rev. 1997 Oct 1;104(4):667-85. Abstract

Myin-Germeys I, van Os J, Schwartz JE, Stone AA, Delespaul PA. Emotional reactivity to daily life stress in psychosis. Arch Gen Psychiatry. 2001 Dec 1;58(12):1137-44. Abstract

Micki Bresnahan, Columbia University
Brian Chiko, Schizophrenia.com
Natalia del Campo, University of Cambridge
Angela Epshtein, Schizophrenia Research Forum
Paul Fearon, Trinity College of Dublin, Ireland
Carla Gallo, Universidad Peruana Cayetano Heredia
Jan Golembiewski, University of Sydney
James Kirkbride, University of Cambridge
Heather Leutwyler, University of California, San Francisco
Hakon Heimer, Schizophrenia Research Forum
Dana March, Columbia University
Maureen Martin, University of California at Irvine
Huan Ngo, Northwestern University
Sami Omer
Kazuko Sakata, University of Tennessee Health Science Center
Linda Scoriels, University of Cambridge
Jean Paul Selten, Utrecht University
Amresh Shrivastava, University of Western Ontario
Nico Stanculescu, Schizophrenia Research Forum
Wim Veling, Parnassia Psychiatric Institute
Victoria Wilcox, Schizophrenia Research Forum

Hakon Heimer
In our usual informal spirit, I will just say that Dana March is at Columbia University in the United States, James Kirkbride is at the University of Cambridge in the U.K., and Wim Veling is typing at us from Parnassia Psychiatric Center in The Hague, Netherlands. I thank Dana for bringing us this idea and cajoling James and Vim to join us. I'll now turn the floor over to Dana.

Dana March
Good afternoon. Thanks for joining us and welcome to the SRF live discussion. Our discussion today will focus on three important topic areas in the study of social factors in schizophrenia. We will spend about 20 minutes in each domain, and our hope is to engage you in a lively exchange regarding one of the most stimulating frontiers of research in social psychiatry. First, James Kirkbride will kick off the discussion by addressing intriguing patterns of findings across populations and settings. Second, I will move our exchange into issues of exposure conceptualization and relevant timing over the life course. Finally, Wim Veling will lead us downstream into a discussion of biological mechanisms (e.g., social defeat, gene-environment interactions, and epigenetics).

James Kirkbride
Hello everybody, welcome along—great to see so many people here for the discussion.

I’m going to kick things off with a brief recap of the main epidemiology findings on schizophrenia and other psychoses. This will focus on how these patterns vary (or don’t) between populations and settings. They were mostly covered in the background reading, but I’ll put them in simple numbered form so that we’re all on the same page. Then I’ll ask some questions to kick off the discussion about how we might best proceed in order to reach a better understanding of “social” and environmental risk factors for schizophrenia. Okay, here’s a recap of what the epidemiology tells us:

1. There is variation in the incidence of schizophrenia across the globe. Rates vary between countries, within countries (increasing with urbanicity, including urban birth), and within small areas (even, e.g., within cities—see Faris and Dunham, 1939—and subsequent contemporary followers).

3. Higher rates of schizophrenia are associated with more marked inequalities in social deprivation—in terms of neighborhood poverty to an extent, but also in terms of social disorganization. Areas with less cohesiveness seem to have higher rates.

4. Rates of all psychotic disorders (schizophrenia and bipolar disorder, e.g.) are increased for immigrants and their offspring (so-called second-generation groups; see SRF related news story and SRF news story).

5. This has been found in the U.K., Netherlands, Sweden, Denmark, and other European countries, and the U.S. (Bresnahan et al., 2007).

6. Raised rates appear highest in immigrants and their offspring when skin color differences between the “background” and “immigrant” populations are greatest—in the U.K., for example, incidence rates are around five to eight times higher in black Caribbean and African immigrants and their offspring than in the white British group, three to four times higher for Asian women (not men), and twice as high for non-British white migrants.

7. The risk of schizophrenia in immigrants and their offspring is not explained by: (a) Selective migration (Selten et al., 2002), (b) higher rates in country of origin, (c) misdiagnosis, or (d) sociodemographic confounds such as age, sex, and ethnicity.

8. Ethnic density phenomenon—the risk of schizophrenia increases in immigrants and their offspring when they live farther away from people of the same ethnic group (i.e., when they are more isolated from their own ethnic group).

Jan Golembiewski
Anyone, is there a known correlation between schizophrenia incidence and a highly stratified society?

Dana March
Jan, there doesn't seem to be a social gradient, if that's what you mean. It seems that social adversity seems to matter most in early life.

James Kirkbride
Jan, that’s a good question. When we see greater social fragmentation, there appear to be great rates of schizophrenia—see Allardyce's work (Allardyce et al., 2005) or some of my own (e.g., Kirkbride et al., 2007). But not so much on socioeconomic lines, though.

James Kirkbride
Perhaps I can suggest a more general point for discussion to kick things off? One thing that strikes me as a good starting point for discussion is this:

We suspect that there is likely to be interaction between genes and environments (Wim Veling will discuss this in more detail later in the session). At present, our biggest, most replicable environmental “risk factors” are urbanicity and migration/ethnicity. Urbanicity and migration/ethnicity can only be markers for increased risk due to other underlying social variables/constructs. How do we, as academic researchers, go about getting past urbanicity and migration to more specific risk factors?

Paul Fearon
James, concentrate on the differences in trying to untangle the underlying causes—for example, why variation in schizophrenia and not bipolar disorder? Why greater rates in Africans and African Caribbeans than Asians, etc.?

James Kirkbride
Paul, I agree. We shall come to that distinction between schizophrenia and bipolar disorder.

Dana March
James, I wonder if it is useful to think about mediators of these effects in terms of two broad classes of the physical (e.g., infections, toxins) and the social (crowding, composition).

Jan Golembiewski
James and all, there are conditions when a social group in one country receives little social status, but the same group—with similar social behaviors—receives high status in another. Take Indians, for example. In Australia and most of Africa, they are highly respected; in Europe, they aren’t. In the U.S., even less.

Dana March
Jan, this is true for black Caribbeans. They have higher social status in the U.S. than in the U.K.

James Kirkbride
Jan, I'm not sure if this is true across the whole of Europe. There have been some issues to do with this in the U.K., but Indian groups are fairly well integrated and respected in the U.K. relative to other groups. The more general point is interesting, though. Relativity seems to hold the key to much social research; context mediates individual risk.

Wim Veling
Dana, about social aspects, Jan may be right that social hierarchy is important.

Jan Golembiewski
Misplacement creates conditions of taboo. Mary Douglas (Douglas, 1966. Purity and Danger. New York: Routledge) posits a general rule that applies equally to inanimate objects as to social dynamics. Bolognaise is clean, acceptable, and welcome in its proper place—that is, on a dinner plate on top of spaghetti—but is not okay on the floor or on my tie. Curiously, misplacement is an intrinsic condition of second-generation immigrants.

Dana March
Wim, Jan, agreed. Social hierarchy may be important, but why? Because of discrimination? How might primate models be useful here?

James Kirkbride
Dana, is there much robust evidence for physical factors (toxins, etc.) at present? We seem to have moved toward a more social hypothesis in recent years.

Dana March
James, I think it may vary across groups and settings. The physical aspects have received far less attention, because data (exposure in utero might be really important) are hard to come by.

James Kirkbride
Dana, I agree. I would suspect that the "physical" aspects of environment would have their greatest effects close to birth (prenatally or in the first few years of life). This, of course, makes them very difficult to study.

Micki Bresnahan
James, maybe it would be useful to discuss what we know about timing.

James Kirkbride
Micki, we will come to timing later in the second half of the debate.

Wim Veling
Dana, there is some preliminary evidence (Pedersen, 2004) that air pollution may be related to schizophrenia incidence.

Dana March
Wim, yes, that's another super difficult exposure to measure. But agreed—intriguing findings.

Paul Fearon
Wim, although pollution and other environmental exposures may be plausible candidates for the increased urban incidence, what physical exposure could account for the markedly raised migrant rates in urban settings?

Jan Golembiewski
Wim, I find the social complexity and lack of integration theory more plausible than air pollution, but I know the pollution theory was explored by David Halpern (Halpern, 1995. Mental Health and the Built Environment. London U.K.: Taylor and Francis).

Sami Omer
Hello everybody. What strikes me is that social deprivation has usually been linked to urbanicity. If deprivation per se had a role, one would expect much higher rates in developing countries.

Jan Golembiewski
Sami, developing countries often (usually) have tight-knit and highly supportive social structures; developed countries don't. In Nigeria, there is no granny dumping; it is a Western phenomenon. The same might be true for those suffering the stress of onset schizophrenia.

Wim Veling
Sami, that also relates to developing countries, because most people share the deprivation.

James Kirkbride
Sami, it's an interesting point. What seems to be true here is that there are possibly different directions to such findings at different levels. For example, take treatment. Treatment with antipsychotics is generally accepted to improve the positive symptoms of psychosis (based on Western medicine), yet there is evidence that course and outcome are better in the developing countries—so what appears to operate in one direction at the individual level may operate in a different direction at the national level.

James Kirkbride
Sami, but I also believe that socioeconomic deprivation per se is probably not a "risk factor" for psychosis. It is probably something more on the lines of relative deprivation or factors correlated with this. The ethnic density effect is one such example. Perhaps there is also a relative deprivation effect?

Sami Omer
Indeed, one would also expect rates of schizophrenia to drop by a significant level with improvement in socioeconomic conditions, similarly to rheumatic heart disease.

Jan Golembiewski
James, one of the ideas I have attempted to study is the variance of texture as visual stimulus, in varying milieu. In some places, white textureless surfaces are normal; in others, they are abnormal. These textures give depleted sensory functions something to process.

Dana March
Okay all, so I would like to start the thread regarding exposure conceptualization and timing of exposure.

We were headed in that direction. With urbanicity, the relevant timing of exposure seems to be at birth (proxy for in utero exposures?), with the effect diminishing markedly after age 15.

Wim Veling
Dana, the same for cannabis: It also has the largest effect when people use it before they are 15. We are working on analyzing age at migration; it might be that those immigrants who migrate at a younger age have higher risk.

Dana March
With the most recent findings on ethnic density (Boydell et al., 2001; Kirkbride et al., 2008; Veling et al., 2008), the exposure was measured at first contact. Is it possible that ethnic density is protective earlier in life, around the transition to adulthood, etc.?

Without systematic data across the life course, it is difficult to know. To understand the relevant timing of exposure would help us eventually to target interventions.

Victoria Wilcox
Dana, speaking of the life course, would it be useful to look at the second peak in incidence in women that occurs later in life and how that might relate to environmental factors?

Dana March
Victoria, that's a really interesting suggestion. Age of onset may be critical in unraveling some of these patterns.

James Kirkbride
Dana, timing around adolescence might be important. This is often a time when we are exploring social contacts, developing peer relationships, and so forth, so perhaps ethnic density at this time has strong effects. I am not sure how it would relate closer to birth.

Dana March
James, that's my hunch as well. I wonder, however, if it might be helpful in buffering the moms from discrimination when they are pregnant, thus bearing on a stress cascade that would otherwise be harmful. We can set up competing hypotheses to figure it out.

Micki Bresnahan
James, adolescence may be the time when the developmental deviation begins to become apparent, even if it was there before.

James Kirkbride
Micki, yes, that's a good point. You are suggesting that these factors have been present from a much earlier period but only become manifest at the time of adolescence? Dana/Wim, this raises the timing issue more generally. Is it likely that the critical timing points for exposure to social factors are around the periods of critical development, both biologically speaking and socially speaking?

Dana March
I agree that we need studies aimed at critical and sensitive periods in social development.

Paul Fearon
James et al., I agree. But I think it’s important to conceptualize this as multiple critical periods (which, socially, might differ between cultural groups).

James Kirkbride
Paul, this is a good point. There is considerable difference by cultural group in terms of the structure of family environments, and such differences may have more pronounced effects in some groups than others on later risk of psychosis—particularly, for example, if the family unit is more fragmented, and there is less familial support around. This is difficult to measure, though, and would require longitudinal data. Does AESOP (Aetiology and Ethnicity of Schizophrenia and Other Psychoses) have much familial data?

Paul Fearon
James, I agree. I was also thinking in terms of age. Regarding family data, they’re limited. Jane Boydell is looking at relatives in the follow-up.

Jan Golembiewski
James, the data exist, naturally as we look from country to country. Surely, it isn't latitude and longitude that create variability.

James Kirkbride
Jan, sorry, I don't understand. I think we can make more use of data available across different settings. Presumably, there is an underlying latent construct that we are trying to measure across these settings even if the specific "exposures" we are attempting to measure are different.

Jan Golembiewski
James, in Australia, the media-projected perception that schizophrenics are evil killers is listed as the number one stigma for sufferers. It occurs late, but it may well have a place in the vicious circle. Dana, we should buffer everyone from discrimination!

Dana March
Jan, agreed. The issue is that discrimination assumes many forms, which is a nice segue into conceptualization of exposure. It's a multilevel concept. At the area level, we deal with both structural and interpersonal discrimination.

Micki Bresnahan
There is a loose use of "social." We have "social" exposures, individual "social" development, and socially structured exposures that may or may not be social. Dana, are you convinced that social development is only shaped by social interaction/factors?

Jan Golembiewski
Micki, I doubt it. Social acceptance is a very complex thing that relates to every aspect of life including eye color.

Dana March
Micki, I think it's also heavily materially influenced (which is a social factor in many senses). I have a broad conception of what social factors are. Perhaps we need better construct refinement and common terms to reshape our research agenda.

Micki Bresnahan
External social factors (that are inherently social in nature) should be different from external factors that are socially structured (e.g., the difference between discrimination and poor housing).

Internal development—developing social skills—that should be shaped by both internal capacities and faculties as well as external exposures (family, peer, society).

Amresh Shrivastava
What really are the constituents of “social factors”? It’s an ill-defined concept that encompasses everything non-biological.

Amresh Shrivastava
Yes, but it can certainly be looked at differently. I call all non-biological factors “ecological” and think these factors may have collective influence.

James Kirkbride
Amresh, agreed. We need to be thinking more carefully about "social" exposures and specifying at which level they occur and when (with a priori reasoning) over the life course they may be relevant.

Dana March
Amresh, Micki, I think it would be productive to have a working group to come to some consensus to set our research agenda in terms of priorities.

Amresh Shrivastava
I think the real name would be “environmental factors”—that means factors like developmental, family, cultural, etc. Yes, we need to think about it clearly.

Wim Veling
Amresh, James, we have plenty of ideas of social exposures (fragmentation, exclusion, family structures, etc.). We only need to explain properly what we mean.

Micki Bresnahan
If you think of social development qua autism, the necessity of making these distinctions becomes clear.

Dana March
Micki, yes! Also, another related question: Do "social factors" also bear on the distribution of psychotic symptoms, and at what age do they appear? I’m thinking of Kristen Laurens' recent paper (Laurens et al., 2008).

Amresh Shrivastava
Like childhood abuse, humiliation, violence, deprivation, etc., affects neurotransmitters? Don’t you think it needs clear quantification and measurement to make a statement?

James Kirkbride
Wim, presumably, see comment above to Jan, they are tapping into one or more common underlying latent traits—stress response to adverse life events/environments at critical periods.

Micki Bresnahan
If the consequence of early exposures (of any external variety) is on the development of social skills in the individual....

Amresh Shrivastava
How then do we explain that many who suffer similar stressful conditions do not develop psychopathology?

Hakon Heimer
Amresh, I think we can chalk it up to differential genetic make-up plus all environmental exposures up to that point.

Jan Golembiewski
Amresh, could it not be something as simple as thought patterns? And how people as individuals cope or rather break down when exposed to stress?

Dana March
James, Wim, I think we are headed toward talk of mechanisms—but how can we get there when we have such crude exposure measurements? One way to move forward is to take the study of the upstream social pathways as seriously as the biological pathways. And Micki, we meet somewhere in between with individuals and groups.

Wim Veling
All, we should study different environmental aspects at different points in the lifetime. During pregnancy: factors related to brain development; during childhood and adolescence: social development, cannabis, etc. Then we can include both biological and social factors, because we know better which aspect we are studying.

James Kirkbride
Micki, it is possible that there are early life insults, which set you off on a life-course trajectory, which increases the probability of further negative events—possibly leading to processes such as sensitization. In other words, early life insults may lead to a self-perpetuating cascade of ever increasing deleterious environments, thus culmulating in increased risk of psychosis?

Dana March
James, an excellent point. What are the trajectories and turning points?

James Kirkbride
Dana, turning points? Not sure. What about familial and neighborhood-level social support/cohesion as a potential buffer for some individuals against these environments? I also agree with you that we struggle from a theoretical base to define our constructs, which limits our ability to measure them.

Wim Veling
James, exactly. Sensitization of the dopamine system is a useful paradigm.

James Kirkbride
Wim, yes, dopamine sensitization is a candidate mechanism, but there are also others.

Amresh Shrivastava
Yes, but these need to be well established. How do we find out biological correlates of social brain insults?

Wim Veling
All, we are already talking about a mix of genes, environment, biological factors. However, there are some complicated issues:

2. But we know that 80 percent of schizophrenia "is determined" by genes—this includes GxE (gene-environment) interactions. 3. To advance research, we need to think of a priori specified hypotheses, studying well-specified genes in relation to functions, for example, stress-related genes, or genes involved in dopamine metabolism.

James Kirkbride
Micki, Wim, me, too—particularly to start with as it includes GxE. It must contain an element of E.

James Kirkbride
Amresh, it is surely a possibility. Neighborhoods with less social disorganization show lower rates of schizophrenia.

Dana March
James, in other words, are some people on the path to psychosis, and are prevented from reaching that point, or vice versa; something sets them off course developmentally? I think the old distinction of predisposing/precipitating is useful here.

Jan Golembiewski
Amresh, from what I understand all the components that make up the aetiology of schizophrenia are found in normal controls. Even voices are heard by 10-15 percent of normal people. It is how this escalates that seems to be problematic—and these “patterns” are likely set up in early life.

Amresh Shrivastava
Do we have any data on gene expression under social stress? Is there a common pathway for all social stress which affects brain biology?

Micki Bresnahan
It may be that studying a metabolic pathway would be more useful. That allows for multiple genes to influence the same outcome.

Dana March
Wim, do you feel like we are moving down the functional road, as opposed to "finding the gene for schizophrenia"?

Wim Veling
Dana, yes, we should focus more on function, not on the Holy Grail of the cause of schizophrenia.

Paul Fearon
Wim, I agree. I'd go as far as to say that with so many different potential exposures, we may never find the “cause” of schizophrenia!

Victoria Wilcox
Amresh, do we even know that social stress is the kind of stress that matters for schizophrenia? Could it be physical stress (e.g., malnutrition, lack of health care) that's responsible for variations in schizophrenia incidence across time, place, and group?

James Kirkbride
Dana, yes, agreed; the distinction between precipitating and predisposing is key. There are developmental trajectories (though subtle) (see Jones et al., 1993; Jones et al., 1994), but presumably, I agree; there must exist "turning points," which could steer you toward or away from schizophrenia—such as the decision to start smoking pot.

Amresh Shrivastava
Maybe yes, you are right. The data are always overstretched in this context.

Wim Veling
Amresh, there are examples from animal research that relate social stress to changes in dopamine system as in schizophrenia patients.

Amresh Shrivastava
I suggest we have an e-mail group to continue our discussion. Bye for now.

Dana March
Wim, whew! How do we join forces productively with geneticists? The E is usually e in GxE investigations. Paul, Wim, we could take a note from Geoffrey Rose here and the folks who study social inequalities as well: The causes of cases are not necessarily the causes of rates.

James Kirkbride
Dana, I think there are increasing numbers of groups of G researchers taking E more seriously. The failure of large main effects of G has helped in this regard. I agree with Wim that we need very careful thought as to putative GxE—i.e., what is theoretically grounded rather than just launching ourselves in.

Jan Golembiewski
Paul, Dr. Humphrey Osmond (Osmond, 1966. Some Psychiatric Aspects of Design. In LB Holland [Ed.], Who Designs America? Garden City, NY: Anchor, pp. 281-318) and ET Hall (Hall, 1975. Mental health research and out of awareness cultural systems. In L Nader and TW Maretzki [Eds.], Cultural Illness and Health. Washington, DC: American Anthropological Association, pp. 97-103) both claimed that schizophrenia was always associated with a retardation of perceptual ability.

Hakon Heimer
James, any way (plans?) to look at genetic variation in genes that code for HPA (hypothalamic-pituitary-adrenal) axis proteins in the AESOP population?

James Kirkbride
Hakon, maybe this is a question better targeted at Paul; he is involved in the AESOP follow-up. I don't know the answer, I'm afraid!

Wim Veling
Paul, no, but we should use the genetic knowledge we have to go further. For instance, on stress sensitivity and COMT gene differences, Jim van Os's group has found some interesting things (Simons et al., 2009; van Winkel et al., 2008; Stefanis et al., 2007). Or expression of pro-inflammatory genes, influenced by infectious agents or stress, etc.

James Kirkbride
Wim, do we believe in COMT? It is on shaky ground among most geneticists (though acknowledged, an absence of main effects should not preclude interactions).

Dana March
James, Wim, I'm not a COMT believer, but I could be converted with evidence.

Wim Veling
James, COMT is an example; if that is out of date, we should study BDNF or whatever more convincing candidate.

Dana March
Lauren Ellman at Columbia is doing really interesting work with cortisol and stress during pregnancy in our schizophrenia cohort in the U.S.

Wim Veling
All, my main point is that there are several interesting biological pathways, of which we know part of the biological chain. We should relate that to our environmental knowledge. Geneticists cannot do that.

Paul Fearon
Jan, we are collecting DNA in the AESOP sample. Robin Murray and others are more up to speed regarding what plans they have for it. I know that Carmine Pariatne has an interest in the pituitary and found increased pituitary volume in the AESOP sample.

Hakon Heimer
Paul, Dana, I wondered if genetic variation in the Afro-Caribbean population might show up.

James Kirkbride
Hakon, Paul, an interesting, tantalizing finding which needs careful (very careful) handling, but it’s important to address genetic variation by ancestry.

James Kirkbride
Dana, with care! We have these "threads" of clues, and we should follow them—as is Wim's point more generally; take the genetic clues and think theoretically about which would hold up to interaction with environmental stimuli. Dana, a more fully fledged answer: I am not sure. But we do know that the val/val genotype has a higher prevalence in black Caribbean populations than others. I think some brave funders and researchers are, at some point, going to have to dispassionately address such issues. We are dealing with an incredibly complex set of disorders with complex aetiology, an aetiology that doesn't stop when it meets political sensitivity.

Dana March
James, but potentially—then onto manic psychosis—we could impact policy in a very positive way, if we learn our lessons from history well.

James Kirkbride
Dana, yes. I need to think about this more than time here allows. Paul, thanks. Sorry we didn't get a chance to come back to this: I wonder whether the findings that schizophrenia is associated with urbanicity may be partially confounded by parental cognitive ability. We know cognition differs between schizophrenia and bipolar disorder. Since it is a heritable trait, perhaps this explains drift to poorer areas for schizophrenia but not bipolar disorder.

Dana March
James, you raise a good point about urbanicity and parental cognitive function.

Paul Fearon
James, maybe. It’s certainly plausible. I can’t help but feel (or maybe I just want to feel) that it’s something more fundamental.

James Kirkbride
Paul, how do you mean more fundamental? Do you mean there is a causative role of the social environment? Despite my parental cognitive function hypothesis, I think there is room for both social drift and cause—mutually reinforcing processes?

Dana March
James, Mark Weiser (Weiser et al., 2007) has looked at diminished cognitive function and urbanicity, though not in the parents.

Jan Golembiewski
Veling, James, do these genes have any other more observable expressions other than increased chance at schizophrenia, such as color blindness or short-sightedness?

Wim Veling
Jan, yes, we know that they are involved in regulation of different brain processes.

Hakon Heimer
Paul, unexpected population difference in common gene SNPs (single-nucleotide polymorphisms) of Afro-Caribbeans or Africans in genes that code for possible mediators of "environmental" stress? Or is that too naive an experimental question? All, we're about five minutes from the official end (though the room is open, and we encourage loitering). Any final statements?

Paul Fearon
We didn't get a chance to come to it, but I think James's findings of lack of variation for manic psychosis is fascinating and needs explanation.

Jean Paul Selten
All, I think it would be useful if we remind the scientific community that the heritability index is so misleading. The misinterpretation of this index is the reason why all the money goes to the geneticists. They will not remind the community of this.

Hakon Heimer
All, any ideas for large-scale collaborations that could be initiated/encouraged?

Dana March
Hakon, "Desperately Seeking Cross-national Collaborators!" We should harmonize studies where possible. The NIH (National Institutes of Health) is putting cash into this!

Wim Veling
Hakon, Dana, we need to get consensus about which E and which G to study, to define priorities. Collaboration is desperately needed, indeed!

Jan Golembiewski
James, I find the observation that schizophrenia patients are less likely to develop cancer a really curious one. Any word?

James Kirkbride
Jan, variation exists in a variety of disorders by a variety of sociodemographic markers; such is the rich heterogeneity of populations. I can't comment with specific reference to cancer as I don't know the literature well enough.

Kazuko Sakata
Wim and Hakon, Dr. Sklar's group published the population genetic study of BDNF Val/Met over the world (Petryshen et al., 2009). I think this kind of study is necessary for future studies to determine the genetic factor.

Victoria Wilcox
James, is there any reason to believe that Asians might have a gene that buffers against migration/ethnicity effects?

James Kirkbride
Victoria, no, but I am not convinced the rate in Asians is as low as commonly suggested. Excess in Asian women in Coid's East London sample was five times that of white British women. It could be differential social support rather than genetics that makes the difference—there may be many cultural differences between Asian women and men in the U.K. to explain these findings.

Jan Golembiewski
James, Asian women in London often have extreme stress but often lack the social support they had at “home.”

James Kirkbride
Jan, this is one working hypothesis we should test. It is possible that Asian women lose the social support they had in their country of origin and may occupy a more marginal place in the U.K. Jan, they are more first-generation—see our paper (Coid et al., 2008).

Jan Golembiewski
James, it would be very interesting to see if those women are more first-generation or second. They could have married into migration, so to speak, and live with impossible in-laws.

Dana March
All, we do need to think about how people come to be in places in which they are exposed. In much the same way that we're interested in migration, geographic and social mobility are key.

James Kirkbride
Dana, I think it has to be in parents because of the urbanicity at birth and later risk of schizophrenia effects. I have read the Weiser paper, though.

James Kirkbride
Dana, no, sadly not. I’d like some, though! Do you know of any? It’s possible to do in birth cohorts, I think, but N becomes the issue.

Dana March
All, another important issue to study is the multiple disadvantaged roles occupied by migrants and ethnic minorities.

Paul Fearon
Everyone, I’ve got to go. James, yep, I agree there's room for drift and cause. I’m just hoping there's some cause there, too. Cheers, all!

Hakon Heimer
Dana, James, Wim, thanks so much for preparing the text and leading us.

Dana March
Signing off, all. See you soon in the virtual world or otherwise....



	Comments on Online Discussion



	Comment by: John McGrath, SRF Advisor
	Submitted 4 March 2009	Posted 6 March 2009
	Excellent background document—thanks! I have a couple of quick comments which are really “free associations” based on the target document. 1. We need to think about specificity of outcome, as well as specificity of exposure. With respect to urbanicity, this seems to link to schizophrenia but not bipolar. Why is it so? 2. When it comes to looking at GxE interactions, I keep thinking about the wonderful paper from Irv Gottesman’s group (Turkheimer et al., 2003) about the interaction between socioeconomic status and the heritability of IQ in young children (a nested twin study in a large U.S. birth cohort). Mindful of how misleading the concept of heritability can be in some contexts (McGrath and Selten, 2008), this paper reminds us how optimal versus suboptimal environments can alter the expression of a massively polygenic surface level phenotype like “IQ.” 3. It is critical that we progress our understanding of migrant status as a risk factor. Can we unravel the role of migrant status per se versus ethnicity? Some risk alleles occur at different frequencies when populations are sorted by ancient geographic origin. Will this be important for schizophrenia research? Will some lines of research be considered more “politically correct” than others? References: Turkheimer E, Haley A, Waldron M, D'Onofrio B, Gottesman, II (2003) Socioeconomic status modifies heritability of IQ in young children. Psychol Sci 14: 623-628. Abstract McGrath JJ, Selten JP (2008). Mental Health: don’t overlook environment and its risk factors. Nature 454; 824. Abstract View all comments by John McGrath

	Comment by: Kathryn Abel
	Submitted 4 March 2009	Posted 6 March 2009
	The concept of “social environment” should include maternal environment, i.e., maternal effects on early programming of development. References: Abel KM (2004) Fetal Origins of Schizophrenia: testable hypotheses of genetic and environmental influences. British Journal of Psychiatry 184: 383-386. Abstract St Clair, D., Xu, M., Wang, P., Yu, Y., Fang, Y., Zhang, F., Zheng, X., Gu, N., Feng, G., Sham, P., He, L. (2005). Rates of Adult Schizophrenia Following Prenatal Exposure to the Chinese Famine of 1959-1961. JAMA 294: 557-562. Abstract McClellan, J. M., Susser, E., King, M.-C. (2006). Maternal famine, de novo mutations, and schizophrenia.. JAMA 296: 582-584. Abstract Khashan A, Abel KM., McNamee R, Goertz M, Baker PN, Webb RT, Kenny L and Mortensen PB. (2008) Higher risk of offspring schizophrenia following prenatal exposure to severe life events. Archives of General Psychiatry; 65:146-152. Abstract Khashan AS, McNamee R, Abel KM, Pedersen MG, Webb RT, Kenny L, Mortensen PB, Baker PM. (2008) Reduced infant birthweight consequent upon maternal exposure to severe life events Psychosomatic Medicine 70:688-694. Abstract View all comments by Kathryn Abel

	Comment by: Paul Fearon
	Submitted 5 March 2009	Posted 9 March 2009
	Important and complex questions! The markedly raised rates observed in the vast majority of studies of ethnic minority/migrant groups in several countries are striking. There are two findings that may provide clues to unraveling the likely complex etiology of these phenomena: 1. South Asians in the U.K. have much lower rates of psychosis than Black Caribbeans, Black Africans, and those of mixed race (Fearon et al., 2006; Coid et al., 2008), yet migrated at broadly similar times and are both visible members of ethnic minority groups. Why is this? 2. I'm not sure if it’s published yet, but hasn't Ezra Susser's group found that “migrant” groups in Israel do not have the raised rates seen in migrants to other countries? If the two points above are true, one possible explanation is the importance of one's expectations and experiences (whether first- or second-generation) in the host country. Unraveling the exact nature of this is the challenge. As the discussion will be taking place on St. Patrick's Day (!), its just possible that I'll not take part live, but will look at the transcript with great interest. References: Fearon P, Kirkbride JB, Morgan C, Dazzan P, Morgan K, Lloyd T, Hutchinson G, Tarrant J, Fung A, Holloway J, Mallett R, Harrison G, Leff J, Jones PB, Murray RM (2006) ‘Incidence of schizophrenia and other psychoses in ethnic minority groups: results from the MRC AESOP study.’ Psychological Medicine 36(11): 1541-1550. Abstract Coid JW. Kirkbride JB. Barker D. Cowden F. Stamps R. Yang M. Jones PB. Raised incidence rates of all psychoses among migrant groups: findings from the East London first episode psychosis study. [Journal Article. Research Support, Non-U.S. Gov't] Archives of General Psychiatry. 65(11):1250-8, 2008 Nov. Abstract View all comments by Paul Fearon

	Comment by: Erika Cilengir
	Submitted 4 March 2009	Posted 13 March 2009
	Regarding immigrant populations, are you looking at the occurrence of early-onset schizophrenia vs. late-onset schizophrenia? I am wondering whether a person who may be predisposed genetically to developing schizophrenia might actually develop the illness once he/she immigrates to a new country and is faced with some of the challenges you describe above. View all comments by Erika Cilengir

	Comment by: Jan Golembiewski
	Submitted 4 March 2009	Posted 13 March 2009
	E.T. Hall, supported by Dr. H. Searles, posits that psychotic episodes are more likely to occur and are more intense when perceptual systems are under stress, particularly because of lack of stimulus. Osmond (1957) and others point out that we all are prone to psychotic episodes if out perceptual systems are sufficiently blinkered. Architecture itself can assist in maintaining normal perceptual functions. Salutogenics (Aaron Antonovsky, 1987) provides clues. Whilst this theory is useful for architecture, it is probably even better suited for the addressing questions about psychosocial implications of urban settlement patterns and urban design. References: Antonovsky, A. (1987) Unravelling the Mystery of Health, San Francisco, Jossey-Bass Inc. Hall, E. T. (1975) Mental Health Research and out of Awareness Cultural Systems. IN MARETZKI, T. W. & NADER, L. (Eds.) Cultural Illness and Health. Washington DC, American Anthropological Association. HALL, E. T. (1990) The hidden dimension, New York, Anchor Books. OSMOND, D. H. (1958) The Seclusion Room-Cell or Sanctuary? Ment Hosp, 9, 18-19. SEARLES, Dr. H. F. (1960) The Non-human Environment, New York, International Universities Press. SEARLES, Dr. H. F. (1965) Collected Papers on Schizophrenia and Related Subjects, London, Maresfield Library. View all comments by Jan Golembiewski

	Comment by: David Yates
	Submitted 5 March 2009	Posted 13 March 2009
	Is selective migration not accepted anymore? Is it not the case that families that migrate are unsettled where they were, and that a reason for some having a bigger proportion of schizophrenia, might be that they are people who do not—cannot—fit in and are therefore more likely to be families that are affected by schizophrenia, and selected for to be more likely to develop the condition when in unfamiliar surroundings, and continue the susceptibility in the progeny? The same will apply to those moving from rural to urban environments. Their progeny will carry a higher genetic risk than the indigenous around them. The parental lack of the cultural experience and familiarity with the local mores and customs in their new placements will continue to have upbringing difficulties which their progeny will be influenced by, and will make the progeny more prone to suffer social ”sensitivity,” less likely to enter or accept local services, and more likely to end up with the more florid and recognizable expressions of schizophrenia. View all comments by David Yates

	Comment by: Jan Golembiewski
	Submitted 14 March 2009	Posted 16 March 2009
	John McGrath’s observation that a “generation of researchers has been inoculated with false beliefs, which has led to an ideological resistance to data….” (2006) is significant because there has been a tendency to disregard social and environmental patterns when looking at schizophrenia, but this politically correct blindness has not been ubiquitous. David Halpern wrote a very worthy book on mental illness in the community (1995) and the introduction of this book includes a literature review on variation in mental illness generally. Halpern reviews a number of papers on the question raised by David Yates about the effect of drift (migration) vs. environmental and social causation and finds that there can be little doubt that both effects are true to some extent. He also points out that migration isn’t a one-way flow. Just as people will move to city cores to find cheaper and more anonymous accommodation, people will also move out of the cities to find respite from the stresses of city life (Halpern, 1995). Having said this, skeptics should note that all psychosis is even more common among second-generation immigrants than the first (Coid et al., 2008), that is, those who did not immigrate because they were already social outcasts as proposed by Yates and others. For this reason we cannot understate the significance of environmental effects in the etiology of any psychosis. I would like to point out another seminal text which has relevance here. Mary Douglas’s Purity and Danger outlines how misplacement creates conditions of taboo. She posits a general rule that applies equally to inanimate objects as to social dynamics. Bolognaise is clean, acceptable, and welcome in its proper place, that is, on a dinner plate on top of spaghetti, but is not okay on the floor or on my tie. Curiously, misplacement is an intrinsic condition of second-generation immigrants; the first generation exercised choice in immigrating, but their children are neither immigrants (from elsewhere) nor natives (from here). They are a social anomaly and they are likely to suffer from social estrangement unless they make a very particular effort to integrate with wider society. Dr. Searles (1960, 1965) posits that social alienation is a correlative of schizophrenia and very likely part of its etiology (thereby conferring with Selten and Cantor-Graae [2005]). This is where the normal applications of salutogenic theory (Antonovsky, 1987) come into play. Salutogenic theory states that better health results from a state of mind which has a fortified sense of coherence. According to the theory, a sense of coherence is fostered by a patient’s ability to find meaning, to comprehend the environment (comprehensibility), and to be effective in his or her actions (manageability). In the case of schizophrenia, all of these abilities fail, making a vicious cycle of cause and effect. I argue that just as the environment may trigger the illness, it can also make up part of the cure (Golembiewski, unpublished, 2009). References: Antonovsky, A. (1987) Unravelling the Mystery of Health, San Francisco, Jossey-Bass Inc. Coid, J. W., Kirkbride, J. B., Barker, D., Cowden, F., Stamps, R., Yang, M. and Jones, P. B. (2008) Raised Incidence Rates of All Psychoses Among Migrant Groups; Findings From the East London First Episode Psychosis Study. Arch Gen Psychiatry, 65, 1250-1258. Abstract Halpern, D. (1995) Mental Health and the Built Environment; More Than Bricks and Mortar, London UK., Taylor and Francis. McGrath, J. (2006) Variations in the incidence of Schizophrenia; Data Versus Dogma. Schizophrenia Bulletin, 32, 195-197. Abstract Searles, D. H. F. (1960) The Non-human Environment, New York, International Universities Press. Searles, D. H. F. (1965) Collected Papers on Schizophrenia and Related Subjects, London, Maresfield Library. Selten, J. P. and Cantor-Graae (2005) Social Defeat: risk factor for schizophrenia? British Journal of Psychiatry, 187, (Editorial) 101-102. Abstract View all comments by Jan Golembiewski

	Comment by: Amresh Shrivastava
	Submitted 16 March 2009	Posted 16 March 2009
	Health inequalities and social environment are closely interlinked (Warr et al., 2009). It clearly influences not only nature and size of pathology but also the treatment outcomes. That social environment modifies the effect of genes has long been known and is being repeatedly replicated, e.g., 5-HTTLPR genotype on PTSD risk (Wille et al., 2008). It has become widely accepted that the psychotic disorders are endpoints of atypical developmental trajectories indexed by abnormal emotional and cognitive development early in life. However, the role of environmental factors in determining these trajectories has received relatively little attention (Bentall and Fernyhough, 2008). Research in mental health has shown a biopsychosocial model for understanding development of mental disorders. The questions today, though, remain 1. Why do people in similar situations not develop the illness? 2. Why do people in diverse situations develop similar psychopathology? It would not be out of place to call for an "ecological" context of research that encompasses the whole “environment” and which may explain gene-environment interaction, time of expression, and severity of pathology for syndromes. The social, psychological, cultural, familial, developmental events and experiences need to be brought under one umbrella for clear understanding. In that sense the social determinants of illness and biological correlates of social factors both become very pertinent. It is no doubt a huge challenge. We hope that new research on ecological sciences can help to unfold biological precursors in order to develop a clear model of genesis for mental disorders, which will be helpful for feasible preventive strategies. References: Warr D, Feldman P, Tacticos T, Kelaher M. Sources of stress in impoverished neighbourhoods: insights into links between neighbourhood environments and health. Aust N Z J Public Health. 2009 Feb;33(1):25-33. Abstract Wille N, Bettge S, Ravens-Sieberer U; BELLA study group. Risk and protective factors for children's and adolescents' mental health: results of the BELLA study. Eur Child Adolesc Psychiatry. 2008 Dec;17 Suppl 1:133-47. Abstract Bentall RP, Fernyhough C. Social predictors of psychotic experiences: specificity and psychological mechanisms. Schizophr Bull. 2008 Nov;34(6):1012-20. Abstract View all comments by Amresh Shrivastava

	Comment by: Jean-Paul Selten
	Submitted 16 March 2009	Posted 16 March 2009
	Excellent background document and interesting comments. The most important reason for the neglect of environmental factors is the so-called heritability index (Schwartz and Susser, 2006). According to textbooks, the "heritability" of schizophrenia is 85 percent. This figure is severely misleading, but makes geneticists rich. They will not address this fraud; we will have to do it. References: Schwartz M and Susser E. The myth of the heritability index. In Beyond Nature and Nurture in Psychiatry: Genes, Environment and Their Interplay. Eds. James MacCabe, Robin Murray, Peter McGuffin, Padraig Wright, Owen O'Daly. Taylor & Francis, Inc., 2006, pp 223. View all comments by Jean-Paul Selten

	Comment by: Alberto Arregui
	Submitted 16 March 2009	Posted 16 March 2009
	Hoping someone has an answer to these two questions: 1. Is early onset schizophrenia (i.e. between 15-24 years of age) still considered a distinct clinical entity? Is it still true that males have a higher prevalence in this age group? 2. When I participated in postmortem studies in Cambridge 1979 thru 1981)I was curious as to why the date of birth of schizophrenia patients (as opposed to schizophrenia-like) clustered between November and February. Is this still just an impression or does someone have better data? View all comments by Alberto Arregui

	Comment by: Brian Chiko
	Submitted 16 March 2009	Posted 17 March 2009
	Stress/demethylation/choline and schizophrenia prevention It increasingly seems that stress and DNA demethylation may be a key pathway that links many of the social and psychological factors identified in the summary with increased risk of psychosis. New research also suggests that a diet rich in methyl donors during pregnancy and early childhood may eliminate or greatly reduce risk of schizophrenia and psychosis. I'd appreciate other people's thoughts on this recent research and theories and if others see this as a potentially fruitful path of research. The supporting research includes recent research by J. David Sweatt’s group at the University of Alabama at Birmingham which has found that with the BDNF gene, demethylation happens rapidly under certain conditions, such as when people experience stress. This demethylation seems to upregulate gene expression. Prenatal supplementation with vitamins rich in methyl donors such as choline is seems to mitigate the effect of prenatal stress on brain development, and greatly lower risk of schizophrenia (tested in rats, and currently in clinical trials at U. Denver on humans). References: Roth TL, Lubin FD, Funk AJ, Sweatt JD. Lasting Epigenetic Influence of Early-Life Adversity on the BDNF Gene. Biol Psychiatry. 2009 Jan 14; Abstract Lubin FD, Roth TL, Sweatt JD. Epigenetic regulation of BDNF gene transcription in the consolidation of fear memory. J Neurosci. 2008 Oct 15;28(42):10576-86. Abstract Stevens KE, Adams CE, Yonchek J, Hickel C, Danielson J, Kisley MA. Permanent improvement in deficient sensory inhibition in DBA/2 mice with increased perinatal choline. Psychopharmacology (Berl). 2008 Jun 1;198(3):413-20. Abstract Zeisel SH. The fetal origins of memory: the role of dietary choline in optimal brain development. J Pediatr. 2006 Nov 1;149(5 Suppl):S131-6. Abstract View all comments by Brian Chiko

	Comment by: German Torres, Diana Ayubcha, Sherry M. Zakhary
	Submitted 16 March 2009	Posted 17 March 2009
	As indicated by March, Kirkbride, and Veling, several studies suggest that people born in urban areas are more susceptible to develop psychoses than individuals born in less densely populated areas. Urbanicity may just be a risk factor that when combined with a particular genetic predisposition, triggers a striking transformation in behavior. It is conceivable that susceptible individuals to psychoses are born with a "social" phenotype of inconspicuous solitary preference, including avoidance behavior and nighttime spatial living. This "social" preference would work relatively well under a solitary phase or low population density. However, when this particular "social" phenotype is placed in a high population density, frequent social contact (e.g., mechanical/touch, olfactory, and/or visual) would elicit the release of abnormal levels of dopamine, serotonin, and/or NPY in specific brain circuits. It should be noted that the above neurochemicals have already been associated with promoting gregarious behavior and aggregation (e.g., socialization) in invertebrate animals (Edwards et al., 2002; de Bono and Bargmann, 1998). Understanding how degraded surroundings (high population density) can degrade behavior (psychosis) in some individuals might refine the social hypothesis of schizophrenia. References: Edwards DH, Yeh SR, Musolf BE, Antonsen BL, Krasne FB. Metamodulation of the crayfish escape circuit. Brain Behav Evol. 2002 Jan 1;60(6):360-9. Abstract de Bono M, Bargmann CI. Natural variation in a neuropeptide Y receptor homolog modifies social behavior and food response in C. elegans. Cell. 1998 Sep 4;94(5):679-89. Abstract View all comments by German Torres View all comments by Diana Ayubcha View all comments by Sherry M. Zakhary

	Comment by: Vera A. Morgan
	Submitted 16 March 2009	Posted 17 March 2009
	Critical to our understanding of the role of adverse social environments as a risk factor for schizophrenia is our capacity to operationalize the concept. This means both identifying specific and valid proxies for adversity exposure, as well as including the timing of the exposure into our studies, whether they be clinical or register-based studies. It also requires some thought on the question of resilience. Who is vulnerable and who is not—and why? Discussions such as this play an important role in developing our understanding of the concepts. View all comments by Vera A. Morgan

	Comment by: Kiumars Lalezarzadeh
	Submitted 17 March 2009	Posted 17 March 2009
	The social environment of many immigrants can entail sociopolitical trauma affecting both national identification and mental well-being. Muldoon, Schmid, and Downes (2008) showed that the negative impact of direct political violence or conflict on mental well-being is mediated by national identification. When immigrants perceive discrimination they can lose their national identification to the new nation to which they have immigrated and develop hostile attitude towards the members of dominant out-group. There is gradual effect of discrimination and perceived discrimination on mental well-being (Jasinskaja-Lahti et al., 2008). 1. To what extent is the above gradual effect on mental well-being transmitted to the second-generation immigrants? Adjustment to immigration has stages going from over-identifying and conforming to the dominant out-group to then returning to one's own in-group up to 28 years after immigration. In the conflict between going from the out-group identification back to in-group identification, the second-generation is developing or being born. During the conformity stage of immigration, self-biasing and self-stereotyping oneself with both positive and negative personality traits of one’s own in-group serves to maintain mental well-being. Even if one is prejudiced and the person's mental well-being is affected, the return to minority in-group identification (self-stereotyping) increases mental well-being (Latrofa et al., 2008). The above is shown to cause multiple states of mental well-being and not being well. 2. How does the conflict going from first identifying with one’s own group (in-group), to conformity and identification with out-group (nation of immigration) and then back to one’s own in-group affect the second generation's mental well-being? References: Muldoon, O. T., Schmid, K. and Downes, C. (2008). Political Violence and Psychological Well-Being: The Role of Social Identity. Applied Psychology, 58(1), 129-145. Jasinskaja-Lahti, I., Liebkind, K. and Solheim, E. (2008). To Identify or Not To Identify? National Disidentification as an Alternative Reaction to Perceived Ethnic Discrimination. Applied Psychology, 58(1), 105-128. Latrofa, M., Vaes, J., Pastore, M. and Cadinu, M. (2008). "United We Stand, Divided We Fall"! The Protective Function of Self-Stereotyping for Stigmatised Members' Psychological Well-Being. Applied Psychology, 58(1), 84-104. View all comments by Kiumars Lalezarzadeh

	Comment by: Karl-Ludvig Reichelt (Disclosure)
	Submitted 12 March 2009	Posted 17 March 2009
	Immigrants and nutrition as possible cause for increase in schizophrenia Most immigrants to Europe come from countries where rice and maize are staple foods. Usually very little milk is used, and many Asiatic and African peoples lack lactase after the age of six or seven. Quite a number suffer from intestinal, endocrine, and behavioral problems (Wändell and Gåfvels, 2007; Meyer et al., 2004; McKenzie et al., 2003; Thabit et al., 2008). Immigrants generally move to obtain better living possibilities, and with low incomes are dependent on grains (bread) and milk products. Since we have found peptides derived from food proteins increased in schizophrenia and autism (Reichelt et al., 1996; Reichelt and Knivsberg, 2003) with confirmed relevant bioactivities (Hole et al., 1979; Drysdale et al., 1982; Cade et al., 2000; Lindström et al., 1986; Sun and Cade, 1999; Sun et al., 1999), it would be expected that immigrant populations should suffer higher rates of some mental disorders (Lorenz, 1990). This pertains also to autism according to Gillberg and Gillberg (Gillberg and Gillberg, 1996). Given that Muslim women cover themselves up, decreases in their D vitamin levels must be usual in most of Europe. For these reasons, nutritional aspects should not be forgotten. The changes in biochemistry are really quite profound. References: Wändell PE, Gåfvels C. High prevalence of diabetes among immigrants from non-European countries in Sweden. Prim Care Diabetes . 2007 Feb 1 ; 1(1):13-6. Abstract Meyer HE, Falch JA, Søgaard AJ, Haug E. Vitamin D deficiency and secondary hyperparathyroidism and the association with bone mineral density in persons with Pakistani and Norwegian background living in Oslo, Norway, The Oslo Health Study. Bone . 2004 Aug 1 ; 35(2):412-7. Abstract McKenzie K, Serfaty M, Crawford M. Suicide in ethnic minority groups. Br J Psychiatry . 2003 Aug 1 ; 183():100-1. Abstract Thabit H, Martin G, Brema I, Daly M, Walsh S, Mannion C, Nolan JJ. Immigrant patients with type 2 diabetes mellitus have poorer initial and on-going glycemic control than a matched population of Irish patients. Ir Med J . 2008 Jun 1 ; 101(6):177-80. Abstract Reichelt KL, Seim AR, Reichelt WH. Could schizophrenia be reasonably explained by Dohan's hypothesis on genetic interaction with a dietary peptide overload? Prog Neuropsychopharmacol Biol Psychiatry . 1996 Oct 1 ; 20(7):1083-114. Abstract Reichelt KL, Knivsberg AM. Can the pathophysiology of autism be explained by the nature of the discovered urine peptides? Nutr Neurosci . 2003 Feb 1 ; 6(1):19-28. Abstract Hole K, Bergslien H, Jørgensen HA, Berge OG, Reichelt KL, Trygstad OE. A peptide-containing fraction in the urine of schizophrenic patients which stimulates opiate receptors and inhibits dopamine uptake. Neuroscience . 1979 Jan 1 ; 4(12):1883-93. Abstract Drysdale A, Deacon R, Lewis P, Olley J, Electricwala A, Sherwood R. A peptide-containing fraction of plasma from schizophrenic patients which binds to opiate receptors and induces hyper-reactivity in rats. Neuroscience . 1982 Jun 1 ; 7(6):1567-73. Abstract Cade RJ et al. (2000) Nutr.Neurosci. 2:57-72. Lindström LH et al. ( 1986) Psychiat Res.19: 93-101. Sun Z and Cade RJ (1999) Autism 3: 85-95. Sun Z et al. (1999) Autism 3: 67-83. Lorenz K (1990) Adv Cereal Sci. Technol.X: 435-469. Gillberg IC, Gillberg C. Autism in immigrants: a population-based study from Swedish rural and urban areas. J Intellect Disabil Res . 1996 Feb 1 ; 40 ( Pt 1)():24-31. Abstract View all comments by Karl-Ludvig Reichelt

	Comment by: Peter Jones
	Submitted 18 March 2009	Posted 19 March 2009
	I enjoyed the discussion and am sorry I missed the live interaction yesterday. I'm left with a couple of questions: Why and how are stressful environments (of the type that are relevant to the urbanicity and migrant:host increased risk) stressful? More to the point, how and why are these social environments stressful to some people and not to others? This may be the relevant diathesis rather than anything proximally related to psychosis. It is true that the vast majority of people who are exposed to these "environments" don't develop psychosis—but do they find them as stressful as people who do? If they don't, then presumably they are incorrectly classified as having been exposed. We need to understand more about the afferent arm of the stress response and GxE mechanisms, rather than the efferent genetic component. The methylation comments were relevant here, but how do social environments lead to such changes? I was at a wonderful Keystone Symposium on schizophrenia and BPD genetics recently where the topical question from the multi-millionaire GWAS scientists was "where's the missing heritability?" It seemed quite a surprise to some that some of it may be in the environment. View all comments by Peter Jones

Live Discussion: The Role of the Social Environment in Psychiatric Research: Outstanding Challenges and Future Directions