The results show that, in this large cross-national sample, prevalence rates of psychotic symptoms are in general smaller than those previously reported in the literature for the United States32
and for Europe8,15,33
, but there is a very wide range in the prevalence, with several countries with an extremely high percentage of subjects reporting at least one symptom (eg, 45.8% in Nepal). Most previous studies, however, have not compared different countries, and, when that has been done, marked differences have been found in specific symptoms but not as widely in the prevalence of disorders.13,34
In a recent meta-analysis, differences in prevalence rates were found between 46 countries according to the economic level, in the opposite direction found here: higher rates in developed countries.35
Also, when comparing ethnic groups within the same area, clear differences in specific symptoms can be observed.15
Here, we have also found differences between countries in the lifetime prevalence of schizophrenia, but the range is clearly smaller than for specific symptoms. Differences between countries, perhaps due to cultural differences, may be a potential cause for concern regarding the robustness of the results. However, those differences are in the proportion of subjects reporting the symptoms, and, in the linear regression, dummies for countries were introduced to control for this potential bias. Furthermore, we found that the impact of symptoms remained the same in separate analyses for each country. Moreover, we also performed the analyses removing the outlier countries, and the results remained essentially the same. Thus, it seems that the variations in the reporting rates across countries do not detract from the relationship between the presence of psychotic symptoms and decreases in functioning.
The results of this study indicate that psychotic symptoms are clearly related with worse functioning even among those who do not meet the diagnostic criteria for a disorder. The relationship is continuous with a significant change in the impact between the absence of symptoms and the presence of at least one symptom. That is, the impact adopts the form of a linear relationship when at least one symptom is present: the more symptoms present, the more the impact on health. There is a clear separation in impact in the threshold between subjects without symptoms and all groups of subjects with symptoms. Subsequently, a linear and small progressive decrease in health status occurs as the number of symptoms increases. This effect is present even when controlling for a lifetime diagnosis of schizophrenia, current treatment of schizophrenia symptoms, presence of comorbid depression, alcohol consumption, socioeconomic status, and other potential confounders.
The results suggest a similar impact of the different psychotic symptoms. These results seem to provide support to the idea of a continuum in psychotic symptoms when at least one symptom is present, with a clear discontinuity from no symptoms to at least one symptom and then a small increase in the impact on health as the number of symptoms increases. Our results point to 2 categorical groups in the general population, one without symptoms and the other with a continuous distribution of psychotic symptoms. Nevertheless, in the current discussion regarding the developments of DSM-V and ICD-11, different authors have warned about the risks of taking dimensional approaches or extending excessively the number of diagnostic categories.36,37
The finding of the marked effect on health status of reporting even one psychotic symptom seems to point out to a categorical change, but that by itself does not imply the threshold of a natural category. Given that not all persons who report a psychotic symptom have a diagnosis of a psychotic disorder, while they are accompanied by significant decrements in health, the identification of such psychotic symptoms should signal the need for further assessment. This would be particularly true in clinical settings and could lead to an exploration of the need for early intervention in a high-risk population.
It is also interesting to see that, in several countries, the prevalence of, for instance, hallucinations (in the last 12 months) is lower than the expected prevalence of schizophrenia (lifetime) for those countries. This probably reflects that since only noninstitutionalized subjects were included in the sample, the more extreme symptom presentations are excluded. In any case, the high percentage of subjects with a self-reported lifetime diagnosis of schizophrenia who report at least one symptom and the clear difference in the number of symptoms reported for subjects with and without a previous diagnosis indirectly support the validity of the symptoms reported.
Limitations of our study include the lack of data on potential determinants of severity or disability associated with psychotic symptoms, such as the number and frequency of episodes, episode length, age at onset, and episode severity as defined by more extensive ratings of individual additional symptoms. Longitudinal studies are needed in order to analyze the natural history of these syndromes in the general population. The cross-sectional nature of the study also does not allow an inference with regard to the causal link between psychotic symptoms and health status to be addressed. Particularly, the main limitation of this work is in the lack of measurement of negative and cognitive symptoms that may better predict course and outcome in psychotic disorders.38
Besides, the study did not collect information about substance use such as cannabis, which could be a relevant mediator of the relationship between psychotic symptoms and health status. Likewise, there are other factors such as stress or trauma that could be related with psychotic symptoms and poor functioning and were not included in the survey.
As commented above, this study has implications for current approaches to the classification of psychotic disorders. In the context of the debate on the feasibility of addressing dimensions in addition to categorical diagnosis in the review of the current diagnostic systems, this study suggests that, within the dimension of reality distortion or positive symptoms, the more symptoms the subject has, the worse functioning and health. That provides support for the whole set of symptoms for that dimension in order to better characterize the outcome. It also indicates that minimal presentations of psychotic symptoms, although strongly varying across countries, have potential clinical relevance everywhere, and it could be important to detect subjects with these symptoms early, with the goal of further thorough screening of those at higher risk, bearing in mind the danger of stigmatization this could involve and the potential negative effects of treatments in persons with minor presentations of symptoms.39
Within the framework of the current debate about the inclusion of a psychosis risk syndrome in the DSM-V,40–43
it is not clear if the presence of one symptom is indicative or not of an incipient clinical disorder, but it does signal the need for a more detailed evaluation.