Self-reported psychotic symptoms in the general population: results from the longitudinal study of the British National Psychiatric Morbidity Survey
Wiles NJ, Zammit S, Bebbington P, Singleton N, Meltzer H, Lewis G ;
Commented by , 23 Jun 2006
Aim of the study
A number of previous studies have shown that milder forms of psychotic symptoms may be present in the general population, but the variability of the prevalence estimates was high ranging from 1% to 17.5%. A number of methodological issues probably account for this variability. Much fewer longitudinal data on the incidence and on the risk factors for the development of psychotic symptoms are available.
The authors had the opportunity to analyse data of a nationwide survey in Great Britain to examine the incidence and risk factors of psychotic symptoms during an 18 month period in a population based sample.
Method
The authors used data from the 2000 Psychiatric Morbidity Survey conducted on a representative sample of 8580 adults in Great Britain. The participants were screened with the Psychosis Screening Questionnaire, an instrument conducted by lay interviewers. People with psychotic disorders at baseline were excluded from the analysis.
A follow-up examination was made at 18 months. Incidence of psychotic symptoms was defined as one or more positive items on the Psychosis Screening Questionnaire at the follow-up, but not at the baseline examination. Logistic regression was used to examine the association between various baseline factors and incidence of psychotic symptoms at follow up.
Results
At baseline 10.9% of participants reported at least one psychotic symptom. But more important was the incidence of psychotic symptoms at follow-up of those individuals who had not reported psychotic symptoms at baseline. This was the case in 4.4% of the general population. The regression model identified six factors that were independently associated with incidence of psychotic symptoms during the 18 months follow-up: alcohol abuse, smoking, high rate of adverse life events, residence in a rural rather than a city area; having only a small primary support group and neurotic symptoms at baseline.
Dr Leucht's comments
This study is one of the few longitudinal studies of the incidence of psychotic symptoms in the general population. It showed that a small, but a considerable number of individuals (4.4%) develop psychotic symptoms (not schizophrenia!) in an 18 month period. A strength of the study certainly is that it was population based, i.e. it was representative for Great Britain.
Furthermore, the longitudinal approach allowed to assess risk factors which is difficult in cross-sectional approaches with only one assessment. Limitations are the use of lay interviews rather than interviews carried out by psychiatrists. The authors discuss that this probably increased the incidence slightly. They also mention that it cannot be excluded that some of the psychotic symptoms occurred during drug intake or medical illnesses.
The risk factors for psychotic symptoms showed some similarities with risk factors for schizophrenia, but there were also some striking differences. Notably, in contrast to previous studies living in a rural area rather than in a city was found to be a risk factor for the development of psychotic symptoms here. A compelling reason for this unexpected finding is not provided. The risk factors number of life events and alcohol abuse are in line with previous research, while cannabis use was only a significant factor in preliminary univariate analyses.
There are high rates of smoking in people with schizophrenia already before the onset of the disorder. But an epidemiological study of this kind can not rule out whether smoking is a risk factor or an attempt of self-medication of subthreshold symptoms. Neurosis had also been found as a risk factor for the development of schizophrenia in previous studies, while having only a small support group was a new, but plausible finding. It is well possible that social isolation can contribute to the development of disturbed perceptions in the form of psychotic symptoms.