Table 3.
Longitudinal Studies Investigating the Relationship Between Insight and Symptoms and Insight and Readmissions
| Study | Design | Participants | Measures | Results |
| Soskis and Bowers24 | Assessment at discharge, follow-up range 3–7 y | 32 patients with schizophrenia | Insight: Attitude Questionnaire | No significant correlation between number of rehospitalizations and attitude variables. |
| Outcome: number of rehospitalizations | ||||
| McEvoy et al26,58 | Assessment approximately 5 d after admission, after 2 wk and at discharge | 52 patients with DSM-III schizophrenia | Insight: ITAQ | No significant correlation between insight and psychopathology at time 1. At time 2 global severity (r = −.41) and BPRS (r = −.35) and at discharge global severity (r = −.30) correlated significantly with insight. Change in insight showed no significant association to change in psychopathology. Significantly less readmissions among patients with insight. |
| Psychopathology: BPRS and CGI Global Severity Item, records about readmissions and time in hospital | ||||
| McEvoy72 | Assessment in outpatient settings, follow-up at 1 y | 25 patients with DSM-III schizophrenia and schizoaffective disorder | Insight: ITAQ | Insight was not related to exacerbation of symptoms. The 3 patients who were rehospitalized had lower insight scores at baseline (15.3 vs 19.2). |
| Outcome: BPRS, CGI, and hospitalizations | ||||
| Michalakes et al70 | Assessment at admission, after 15 and 30 d, and at discharge | 42 patients with schizophrenia | Insight: ITAQ | Insight improved significantly and was significantly related to BPRS scores only at discharge (r = −.40) |
| Psychopathology: BPRS | ||||
| Kemp and Lambert32 | Assessment at admission and at discharge (3–6 wk after admission) | 29 patients admitted to acute ward and met DSM-III-R criteria for schizophrenia | Insight: modified version of SUMD | Symptoms were significantly correlated with insight subscales at admission and discharge. Improvement in present insight was not associated with change in symptoms. Improved insight into past symptoms was associated with improvement in negative (r = .53) and positive symptoms (r = .41) and general psychopathology (r = .53). |
| Psychopathology: PANSS | ||||
| Jørgensen et al68 | Assessment in first week of admission, reassessment every 2 wk. Total of 5 assessments. | 50 patients with DSM-IV and ICD-10 schizophrenia (and at least one previous admission) | Insight: IS | Patients were divided into 3 categories: total remission of delusional beliefs, partial remission, and no remission. Insight increased in all groups. For the majority of patients (68%), insight was not influenced by the recovery from delusional beliefs. However, of the minority of patients who remained low on insight (22%), all recovered only partially from symptoms. |
| Psychopathology: PANSS | ||||
| Smith et al73 | Assessment at discharge and after a mean of 20.4 (inpatients) and 86.7 d (outpatients) | 33 patients with schizophrenia and schizoaffective disorders (Structural Clinical Interview for DSM-IV) | Insight: SUMD | No significant correlation between insight and positive, negative, and disorganized symptoms at discharge. At follow-up, disorganized symptoms were correlated with current awareness and misattribution. Changes in past awareness were associated with changes in disorganized symptoms but not with changes in positive or negative symptoms. |
| Psychopathology: SANS, SAPS | ||||
| Chen74 | Assessment at admission and at discharge | 59 inpatients with schizophrenia or schizoaffective disorder | Insight: SUMD | Insight was not related to symptoms at time 1. Upon discharge, emotional withdrawal, disorganization, hostility, suspiciousness, unusual thought content, and uncooperativeness were related to poor insight. Reductions in uncooperativeness and emotional withdrawal were correlated with improved insight. |
| Psychopathology: BPRS | ||||
| Caroll et al75 | Assessment before discharge, assignment to intervention or control group, and reassessment after 12 wk | 100 patients with DSM-III-R schizophrenia (with at least one previous psychotic episode) | Insight: ITAQ | At baseline, insight correlated negatively with positive (r = −.41) and negative symptoms (r = −.20). In the stepwise regression final equation, only positive symptoms (ß = .43) and depression scores (ß = .26) made significant contributions to predicting insight. |
| Symptoms: PANSS | Change in positive and negative symptoms did not correlate significantly with change in insight. | |||
| Weiler et al69 | Assessment at admission and discharge from psychiatric hospital | 81 patients (DSM-III-R schizophrenia) | Insight: ITAQ | Change in insight and change in BPRS total correlated significantly (r = .52) as well as insight change and change in positive symptoms (r = .40). Associations were strongest for the disorganized subtype and lowest for the paranoid subtype. |
| Psychopathology: BPRS | ||||
| Yen et al76 | Baseline assessment of outpatients, 1-y follow-up | 74 patients with schizophrenia (DSM-IV) | Insight: SAI, SAI-E | Rehospitalized patients had significant lower scores for insight into treatment at index assessment but not for insight into mental health status, psychotic experiences, or total SAI-E scores. |
| Outcome: frequency of hospitalization | ||||
| Mintz et al77 | Assessment at admission to early psychosis program, after 3, 6, and 12 mo. | 180 patients with first-episode psychosisa | Insight: PANSS insight item (G12) | Participants with low insight had more positive and negative symptoms at all assessments. |
| Psychopathology: PANSS | Stable insight was associated with improved symptoms compared with stable lack of insight or detoriation in insight. Improved insight by 1 y was associated with improved symptoms. | |||
| Gharabawi et al8 | First assessment at enrollment in a risperidone treatment trial, followed by assessments at weeks 12, 24, 48, and 50. | 614 patients with schizophrenia or schizoaffective disorder | Insight: PANSS insight item (G12). | Patients improved significantly on PANSS total from baseline to end point, regardless of baseline insight. Patients with severe lack of insight at baseline had the highest overall level of symptoms at baseline and follow-up. Changes in insight scores corresponded with changes in symptom domains. Correlations were highest for changes in disorganized thought, followed by negative symptoms and anxiety/depressive symptoms. |
| Symptoms: PANSS total and 3 factors: negative symptoms, anxiety/depressive symptoms, and disorganized thought |
Note: Abbreviations are explained in footnote to table 1. DSM-IV, schizophrenia, schizophreniform disorder, delusional disorder, brief psychotic disorder, psychotic disorder not otherwise specified, or drug induced psychosis; CGI, Clinical Global Impressions; DSM-III, Diagnostic and Statistical Manual for Mental Disorders, Third Edition; DSM-III-R, Diagnostic and Statistical Manual for Mental Disorders, Revised Third Edition; DSM-IV, Diagnostic and Statistical Manual for Mental Disorders, Fourth Edition; ICD-10, International Classification of Diseases, 10th Revision; SAI-E, Schedule for Assessing the 3 components of Insight Expanded Version.