This is my first posting in a series which I am planning to do on the subject of the epidemiology of schizophrenia. Epidemiological studies are gradually becoming more sophisticated, less reductionistic and multilevel. To remain credible and to serve a useful function, social, developmental and genetic epidemiology must address phenomena at all levels of import. I have used epidemiological research, as well as neuroscience (particularly the effects of stress/anxiety/fear on the brain) and phenomenological research (including extensive and long-term clinical experience in inpatient and outpatient settings) to formulate my views on severe mental illness which I have variously called a probabilistic epigenesis model or psychoneurodynamic model of the schizophrenias. Personally, in addition to the research on the effects of anxiety/stress/fear on CNS structure and function, I believe that epidemiological studies, as well as phenomenological research, are a much needed corrective to the strikingly reductionistic trends in research on severe mental illness.
Attempts to understand persons with severe mental illness through neuroimaging, psychopharmacology (which continues to confuse the public with its marketing strategies which fail to differentiate causation and correlation), reductionistic neurobiology, etc., without taking into account broad domains of neurobiological inquiry, phenomenological and epidemiological processes are significantly incomplete and can be scientifically misleading. The astronomer Arthur Eddington once told a parable about a man studying deep-sea life using a net with three-inch mesh. After bringing up repeated samples, the man concluded that there are no deep-sea fish smaller than three-inches in length! The physicist Werner Heisenberg cogently and wisely stated: “Science does not describe and explain nature, rather, nature as exposed to our method of questioning.”
Strong evidence has emerged across many studies that the course and outcome of schizophrenia is more benign in the developing countries, which do not have access to complicated psychopharmacological regimens (Hegarty et al 1994 performed a meta-analysis of outcome studies-representing 320 studies and 51,800 subjects-demonstrating that outcomes in schizophrenia at the beginning and end of the last century were roughly comparable-despite the hegemony of psychopharmacological treatment in the last 3-4 decades of that century-outcomes have worsened in the last few decades of the 20th century. The authors attribute this to the use of a more narrow definition of schizophrenia. However, one could also hypothesize that positive outcomes have declined due to psychosocial and sociocultural factors operative in treatment systems as well as in the patient’s wider social environment.).
Because of its importance, the first posting will summarize the research on the geographic course and outcome of schizophrenia, in particular, the comparisons between developing and developed countries. An early study on the island of Mauritius (Murphy & Raman 1971), examining 12-year outcomes of all patients with a diagnosis of schizophrenia on first admission to hospital, discovered that 64% had no symptoms and were independent. The WHO embarked on two large multicenter studies, the International Pilot Study of Schizophrenia (IPSS) and the Ten Country Study to study this issue. The IPSS was conducted simultaneously in nine countries. The three centers reporting the highest proportion of patients with poor overall outcome were in the developed countries (Aarhus, Denmark 48%, London 41% and WashingtonDC 45%). The three centers reporting the highest proportion of patients with good overall outcome were in the developing countries (Agra, India 66%; Cali, Columbia 53%; and Ibadan, Nigeria 86%).
Overall outcome was rated in five categories based on combined criteria in three outcome domains: percentage of follow-up time spent in psychotic episode, presence or absence of social impairment, and type of remission after episodes. Eight of the nine centers went on to participate in a 5 year follow-up study in which the outcome profile in the developing countries remained superior to that observed in the developed countries. About two-thirds of the patients at the Ibadan site and Agra site were asymptomatic at the 5 year follow-up. Longer-term outcomes (10 years or more) were reported in two developing and one developed site, extending the research evidence in favor of better outcomes in the developing countries. Leon (1989) studied 83% of the Cali IPSS patients and demonstrated that clinical outcome at 10 years was good (complete or partial recovery) in 51% of the patients. Dube et al (1984) studied 62 of the 140 Agra patients 13-14 years after inclusion in the research protocol. Sustained remission was found in 46% of patients and 56% were judged to be clinically ‘normal’ at assessment. Carpenter and Strauss (1991) assessed 40% of the original WashingtonDC sample of the IPSS at 11 years and found little change in the functioning of patients (i.e., in terms of social contacts, symptoms, employment, etc).
The Ten Country Study coordinated by the WHO followed the IPSS using twelve centers (Sartorius et al 1986; Jablensky et al 1992). Again, outcomes were more favorable in developing countries (on five of the six measures of best outcome). Further development of this research disparity was demonstrated in the WHO International Study of Schizophrenia (ISoS). This collaborative research project included the IPSS and Ten Country Study cohorts among others. Using the ISoS, Hopper and Wanderling (2000) replicated the developing versus developed differential in the long-term (more than 13 years follow-up) and under ICD-10 diagnostic criteria for schizophrenia. They demonstrated that various biases such as ascertainment and loss to follow-up bias, as well as diagnostic issues, could not account for the differences in outcomes.
Bresnahan, Menzies, Varma and Susser (2003), in their “Geograhical variation in incidence, course and outcome of schizophrenia: a comparison of developing and developed countries” in Robin Murray et al’s edited volume “The Epidemiology of Schizophrenia” published by Cambridge University Press, speculated on the possible explanations for differences in outcome. They postulated that the most commonly proposed explanations fall into four categories: family relationships, informal economics, community cohesion, and segregation of the mentally ill.
These authors noted that in terms of family relationships, in developing countries patients remain in the family’s care more frequently and social integration within the family is not disrupted. Families remain closely involved in treatment and recuperation of their ill family member. In terms of informal economics, in the developing countries reintegration into work roles is the rule rather than the exception. In these countries, informal economies may provide more diverse opportunities for this reintegration to occur (see the work of Richard Warner on the role of employment in the recovery from schizophrenia). In developing countries, patients are less likely to be segregated in institutions than their counterparts in the developed world. Also, community cohesion may be greater in the developing countries.
Breshnahan et al (2003) concluded: “That developing countries better serve those with schizophrenia than developed countries stands in contrast with what is known about the benefit to individuals of modern therapies (e.g., medication...), treatments often unavailable in these settings...The contrasting experiences of developed and developing countries with respect to schizophrenia may be interpreted as providing evidence for socioenvironmental influences in this disorder” (p. 30).
When I was affiliated with RocklandPsychiatricCenter I had the opportunity to interact with some of the researchers involved in this area of study. Kim Hopper, of the NathanKleinPsychiatricResearchCenter, invited Savita Malhotra MD, a research and community psychiatrist from ChandigarhIndia, to present her data and views on this outcome differential between the developing and developed world. She presented socio-cultural factors as the most likely explanations for this disparity, including the importance of spiritual beliefs (in her site, primarily Hindu beliefs on building good karma and reparation) as well as lack of segregation and family and social cohesion.
Brian Koehler PhD
New YorkUniversity
80 East 11th Street #339
New YorkNY10003
212.533.5687
