When I was 9 years old, I developed a passion for collecting postage stamps. Over the next 2 years, I spent many hours amassing thousands of stamps, and sticking them into albums according to the year and the country from which they emanated – pages and pages of bog-standard British stamps followed by fewer pages of beautiful stamps from African countries that had long since disappeared (e.g. Abyssinia).
I suspect that Sandra Matheson and her colleagues (Matheson et al.
Reference Matheson, Shepherd and Carr2014) share some hoarding genes with me as they are obsessed with systematic reviews and meta-analyses in the same way that I was devoted to philately. In this equivalent of my stamp albums, the authors bring together every systematic review pertaining to schizophrenia that they can lay their hands on. Thus, the authors have trawled through almost 1500 systematic reviews in the schizophrenia literature, just as I used to scan through the incoming mail of my relatives and rummage through the old files of any consenting adult.
The reviews come from the ‘Schizophrenia Research Library’ in Sydney, Australia. Matheson and her colleagues assessed all the systematic reviews, graded the evidence on the different topics as of ‘high’ or ‘moderate’ quality (it is not very clear what happened to ‘low’-quality evidence); they then quantified any findings as of small, medium or large effect.
Of course, many of the conclusions are to be expected. No one will be astonished to read that ‘Patients have relatively poor cognitive functioning, and subtle, but diverse structural brain alterations, altered electrophysiological functioning and sleep patterns, minor physical anomalies, neurological soft signs and sensory alterations.’ Similarly, readers will be relieved that high-quality evidence shows that antipsychotics show a medium effect size in reducing symptoms. They may, however, be startled to learn that high-quality evidence also indicates that Ginkgo biloba reduces positive symptoms or that there are medium effect sizes for adjunctive non-steroidal anti-inflammatory drugs in relieving positive symptoms.
It is significant that there is a paucity of replicable findings in functional neuroimaging, given the large amount of money that has been invested in this technique. However, importantly, the excess of pre-synaptic dopamine in the striatum is supported by high-quality evidence showing a large effect size, thus solidifying the claim that dopamine dysregulation is the final common pathway underlying much of psychosis (Howes & Murray, Reference Howes and Murray2013).
Epidemiology does not provide high-quality evidence for findings with large effect size but a lot of such evidence for findings of medium effect, plus large effects from moderate-quality studies. As the authors state: ‘Risk for schizophrenia is increased with cannabis use, pregnancy and birth complications, prenatal exposure to Toxoplasma gondii, childhood central nervous system viral infections, childhood adversities, urbanicity and immigration (first and second generation), particularly in certain ethnic groups. Developmental motor delays and lower intelligence quotient in childhood and adolescence are apparent.’ Only one of these findings is contentious.
The authors state that ‘it is very striking that there are solid positive findings, with large and medium effect sizes, for numerous adjunctive psychosocial treatments’. I do not doubt that this is a true reflection of the literature, and that psychosocial support is vital to patient recovery. But does it mean that all the psychosocial therapies listed here as beneficial have specific as opposed to general effects? I am not so sure, for there is a major confound in many studies of psychosocial interventions. To caricature this, the trials tend to compare an innovative psychosocial intervention provided by an enthusiastic team led by a charismatic 35-year-old clinician with ‘treatment as usual’ provided by a team led by a disillusioned old psychiatrist idling away his time until retirement while dreaming of ways of poisoning his clinical manager. No wonder the innovative psychosocial treatment wins out!
There are, of course, limitations that the authors accept. For example, I used to think that meta-analyses were unbiased. However, my experience as an editor has disabused me of that. When a meta-analysis is published with negative results, researchers who dislike the findings attack the authors for having rigged the inclusion criteria to discard papers with positive findings. They then go on to publish their own meta-analysis in which they exclude some of the negative studies in the original paper but bring back their favourite papers. Not surprisingly, they find much more favourable results! Cognitive behavioural therapy for psychosis is one such arena where repeated meta-analyses carried out by enthusiasts and sceptics conflict.
Another limitation is that there is practically no mention of genetics. The authors do not feel confident to review this field, which is sensible as schizophrenia genetics is moving fast and the conclusions of today may be overturned within a couple of years.
So, congratulations to the authors for their industry, and for the way in which they examined a huge number of systematic reviews and meta-analyses in a standardized way. But should one rush off and prescribe Ginkgo biloba or non-steroidal anti-inflammatories to one's patients? Of course not; rather one should go to the original literature and make up one's own mind. What one can say is that the authors provide a unique compendium of the systematic reviews on schizophrenia. Here one will find references to reviews that will provide one with a good start to reading about any particular fact (or fiction).
Finally, although collecting meta-analyses and postage stamps both appeal to those with obsessional traits, it is clear that the former is more useful than the latter. This meta-analysis of meta-analyses demonstrates that we know a lot about schizophrenia. A minority of the ‘facts’ will jolt the reader. But are they really true or will they go the same way as Abyssinia? Only time and more research will tell.
You have
Access
When I was 9 years old, I developed a passion for collecting postage stamps. Over the next 2 years, I spent many hours amassing thousands of stamps, and sticking them into albums according to the year and the country from which they emanated – pages and pages of bog-standard British stamps followed by fewer pages of beautiful stamps from African countries that had long since disappeared (e.g. Abyssinia).
I suspect that Sandra Matheson and her colleagues (Matheson et al. Reference Matheson, Shepherd and Carr2014) share some hoarding genes with me as they are obsessed with systematic reviews and meta-analyses in the same way that I was devoted to philately. In this equivalent of my stamp albums, the authors bring together every systematic review pertaining to schizophrenia that they can lay their hands on. Thus, the authors have trawled through almost 1500 systematic reviews in the schizophrenia literature, just as I used to scan through the incoming mail of my relatives and rummage through the old files of any consenting adult.
The reviews come from the ‘Schizophrenia Research Library’ in Sydney, Australia. Matheson and her colleagues assessed all the systematic reviews, graded the evidence on the different topics as of ‘high’ or ‘moderate’ quality (it is not very clear what happened to ‘low’-quality evidence); they then quantified any findings as of small, medium or large effect.
Of course, many of the conclusions are to be expected. No one will be astonished to read that ‘Patients have relatively poor cognitive functioning, and subtle, but diverse structural brain alterations, altered electrophysiological functioning and sleep patterns, minor physical anomalies, neurological soft signs and sensory alterations.’ Similarly, readers will be relieved that high-quality evidence shows that antipsychotics show a medium effect size in reducing symptoms. They may, however, be startled to learn that high-quality evidence also indicates that Ginkgo biloba reduces positive symptoms or that there are medium effect sizes for adjunctive non-steroidal anti-inflammatory drugs in relieving positive symptoms.
It is significant that there is a paucity of replicable findings in functional neuroimaging, given the large amount of money that has been invested in this technique. However, importantly, the excess of pre-synaptic dopamine in the striatum is supported by high-quality evidence showing a large effect size, thus solidifying the claim that dopamine dysregulation is the final common pathway underlying much of psychosis (Howes & Murray, Reference Howes and Murray2013).
Epidemiology does not provide high-quality evidence for findings with large effect size but a lot of such evidence for findings of medium effect, plus large effects from moderate-quality studies. As the authors state: ‘Risk for schizophrenia is increased with cannabis use, pregnancy and birth complications, prenatal exposure to Toxoplasma gondii, childhood central nervous system viral infections, childhood adversities, urbanicity and immigration (first and second generation), particularly in certain ethnic groups. Developmental motor delays and lower intelligence quotient in childhood and adolescence are apparent.’ Only one of these findings is contentious.
The authors state that ‘it is very striking that there are solid positive findings, with large and medium effect sizes, for numerous adjunctive psychosocial treatments’. I do not doubt that this is a true reflection of the literature, and that psychosocial support is vital to patient recovery. But does it mean that all the psychosocial therapies listed here as beneficial have specific as opposed to general effects? I am not so sure, for there is a major confound in many studies of psychosocial interventions. To caricature this, the trials tend to compare an innovative psychosocial intervention provided by an enthusiastic team led by a charismatic 35-year-old clinician with ‘treatment as usual’ provided by a team led by a disillusioned old psychiatrist idling away his time until retirement while dreaming of ways of poisoning his clinical manager. No wonder the innovative psychosocial treatment wins out!
There are, of course, limitations that the authors accept. For example, I used to think that meta-analyses were unbiased. However, my experience as an editor has disabused me of that. When a meta-analysis is published with negative results, researchers who dislike the findings attack the authors for having rigged the inclusion criteria to discard papers with positive findings. They then go on to publish their own meta-analysis in which they exclude some of the negative studies in the original paper but bring back their favourite papers. Not surprisingly, they find much more favourable results! Cognitive behavioural therapy for psychosis is one such arena where repeated meta-analyses carried out by enthusiasts and sceptics conflict.
Another limitation is that there is practically no mention of genetics. The authors do not feel confident to review this field, which is sensible as schizophrenia genetics is moving fast and the conclusions of today may be overturned within a couple of years.
So, congratulations to the authors for their industry, and for the way in which they examined a huge number of systematic reviews and meta-analyses in a standardized way. But should one rush off and prescribe Ginkgo biloba or non-steroidal anti-inflammatories to one's patients? Of course not; rather one should go to the original literature and make up one's own mind. What one can say is that the authors provide a unique compendium of the systematic reviews on schizophrenia. Here one will find references to reviews that will provide one with a good start to reading about any particular fact (or fiction).
Finally, although collecting meta-analyses and postage stamps both appeal to those with obsessional traits, it is clear that the former is more useful than the latter. This meta-analysis of meta-analyses demonstrates that we know a lot about schizophrenia. A minority of the ‘facts’ will jolt the reader. But are they really true or will they go the same way as Abyssinia? Only time and more research will tell.
Declaration of Interest
None.