Insight in schizophrenia: a meta-analysis
Mintz AR, Dobson KS and Romney DM;
Commented by , 26 May 2003
Aim of the study
There are estimates that between 50% and 80% of patients with schizophrenia do not believe that they are ill. Or they do accept that they experience symptoms, but they attribute these symptoms to other causes than a mental disorder. This has an important impact on compliance. Many patients may only take medication due to pressure from their families or because they just follow their doctors’ recommendations.
Over the last decades a considerable amount of research in the study of insight in schizophrenia has been done. Current concepts operationalise insight according to five dimensions which comprise the patient's awareness of mental disorder, awareness of the social consequences of disorder, awareness of the need for treatment, awareness of symptoms and attribution of symptoms to disorder.
One of the main areas of interest is the relationship between insight and symptoms. Intuitively one would argue that pronounced symptoms should be associated with poor insight. However, despite the development of psychometrically sound measurement tools, the results from previous studies have been inconclusive. Therefore the authors undertook a systematic review and meta-analysis on the topic.
Method
A MEDLINE and PSYCINFO search was undertaken to identify relevant English-language articles. Effect sizes were calculated for the relationship between insight and global, positive, negative and depressive symptoms. A global measure of insight and the five different components of insight as summarised above were analysed. Furthermore, the effects of a number of moderator variables on the effect sizes such as demographic, clinical and study characteristics were assessed.
Results
40 published studies on the topic could be included in the meta-analysis. There was a small, but statistically significant negative relationship between insight and global, positive and negative symptoms. Furthermore, there was also a small positive relationship between insight and depressive symptoms in schizophrenia.
Significant moderator variables for the relationship between insight and symptom clusters were acute patient status and mean age of onset of the disorder, i.e. acutely ill patients and patients with later age of onset had lower insight. Statistical tests for publication bias did not suggest that any studies had obviously been missed.
Discussion
The results suggest that if patients with schizophrenia have pronounced global, positive and negative symptoms, their degree of insight is low. On the other hand, if patients are very depressed, their insight increases. However, the authors discuss that the effect sizes found were relatively small. They lay in a range of correlation coefficients between r=0.10 to r=0.40 so that symptoms described only 3-7% of the variance.
The main explications are that other factors such as patient status and premorbid functioning may also be involved in insight in schizophrenia. Furthermore, the relationship between symptoms and insight might not be linear, but meta-analyses cannot examine curvilinear relationships.
Recent theories suggest that lack of insight is the consequence of cognitive dysfunction, resulting in the patients’ inability to recognize that they suffer from a disorder. Another concept understands poor insight as a coping strategy where patients are somehow aware of their illness, but deceive themselves to preserve their self-esteem.