SEARCH

SEARCH BY CITATION

Keywords:

  • alcohol use disorder;
  • bipolar disorder;
  • depressive disorder;
  • insight;
  • schizophrenia

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Aims:  To compare the level of insight among six groups of patients with psychiatric disorders, including those with schizophrenia (SZ), bipolar I disorder (BP), or depressive disorders (DD) who had or did not have comorbid alcohol use disorder (AUD).

Methods:  A total of 285 outpatients meeting the aforementioned criteria were recruited into the study. The Schedule of Assessment of Insight–Expanded version (SAI-E) was used to measure subjects' insight. Analysis of covariance (ancova) was used to compare the levels of insight among the six groups of subjects.

Results:  Regardless of whether patients had comorbid AUD or not, patients with DD had higher levels of insight than did patients with SZ. Comorbid AUD had independent effects on the differences in the level of insight between patients with DD and BP and between patients with BP and SZ. No statistically significant difference in insight was found between patients with the same psychiatric diagnosis with and without comorbid AUD.

Conclusions:  In addition to psychotic features and clinical states, comorbid AUD should be taken into consideration when comparing the level of insight among patients with different psychiatric diagnoses.

THE INSIGHT OF individuals with psychiatric disorders generally indicates an awareness of the psychiatric symptoms experienced, the presence of psychiatric disorders, and awareness of the achieved effect of treatment.1,2 Research over the past few decades has found that lack of insight may compromise treatment compliance and clinical outcome for patients with mental disorders.3,4 Insight deficit was found not only in patients with schizophrenia (SZ)3 and bipolar disorder,5 but also in patients with depressive disorders (DD).6 Relationships of insight to diagnosis have also been studied in order to add further clues about differences among various psychiatric disorders.7 For example, several studies have found that insight deficits were more severe in SZ than in major DD with or without psychotic features.1,8–10 The results, however, of the studies comparing the insight between SZ and bipolar disorder and between bipolar disorder and DD have been mixed. While some studies found that patients with bipolar I disorder (BP) had less insight than those with major DD,8,11–14 Ghaemi et al. found that insight was similarly impaired in bipolar and unipolar major DD.15 While some studies found that patients with SZ had less insight than those with psychotic mania or mixed mania,7 other studies found that patients with SZ had insight deficits similar to those with bipolar disorder in the acute phase,8,10,14 and to bipolar disorder with psychotic features.1,16,17

One of the possible explanations for the mixed results of studies comparing insight among patients with various psychiatric diagnoses may be that these studies have not taken comorbid alcohol use disorder (AUD) into consideration. It has been found that high proportions of patients with SZ, bipolar disorder, and DD have comorbid AUD.18 Meanwhile, alcohol has neurotoxic effects that frequently result in significant neurocognitive deficits,19 and impaired neurocognition has been found to be associated with poor insight in patients with SZ20 and BP.21 Thus, it seems reasonable to hypothesize that comorbid AUD has an impact on the level of insight in SZ, bipolar disorder, and DD. To our knowledge, however, no previous studies have compared the level of insight among patients with different psychiatric diagnoses versus presence or absence of comorbid AUD.

The aim of the present study was to compare the level of insight among six groups of patients with psychiatric disorders, including those with SZ, BP, or DD who had or did not have comorbid AUD. We hypothesized that both psychiatric diagnoses and comorbid AUD have effects on the level of insight.

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Participants

Patients from the psychiatric outpatient clinics of two university hospitals and one psychiatric teaching hospital in Kaohsiung City and County in southern Taiwan, who met criteria for the diagnosis of SZ, BP, or DD based on DSM-IV22 using the Structured Clinical Interview (SCID)23 were consecutively invited to receive a screening interview. Subjects who refused to participate or were unable to complete the study interviews due to prominent psychotic symptoms or cognitive deficits were excluded. We screened their drinking habits using the Chinese version24 of the Alcohol Use Disorders Identification Test (AUDIT).25 The AUDIT is composed of 10 questions, including three quantity–frequency questions, three alcohol-related behaviors questions, and four alcohol-related consequences or harm questions. A previous study found that individuals whose total AUDIT score was ≥8 were more likely to have AUD than those whose total AUDIT score was <8.24 Thus, those whose total AUDIT score was ≥8 in the screening were invited to receive a clinical interview by two research psychiatrists to determine the diagnosis of AUD according to the taxonomies of DSM-IV, including alcohol dependence and abuse.22 A total of 529 subjects with SZ, 197 subjects with BP, and 385 subjects with DD received AUDIT screening. Of them, 51 subjects with SZ, 19 subjects with BP, and 67 subjects with DD were determined to have comorbid AUD based on the AUDIT screening and the clinical interviews. Meanwhile, one-seventh of subjects whose AUDIT score <8 were randomly selected as the comparison non-comorbid group. Overall, 67 subjects with SZ, 40 subjects with BP, and 41 subjects with DD were selected as the non-comorbid groups.

Measures

Schedule of assessment of insight–expanded version

The Schedule of Assessment of Insight–Expanded version (SAI-E) measures multiple dimensions of insight, including compliance with treatment, recognition of illness, re-labeling of psychotic phenomena, and awareness of changes in mental functioning and psychosocial consequences of the illness.26 The maximum score of the SAI-E is 24, with higher SAI-E scores indicating greater insight. The Cronbach alpha for the SAI-E was 0.96, and inter-rater reliability (r) between the two research psychiatrists was 0.86 in the present study.

Brief psychiatric rating scale

We used the 16-item Brief Psychiatric Rating Scale (BPRS) to measure major psychotic and non-psychotic symptoms in participants, on a 0–6 scale.27 Higher total BPRS score indicates more severe psychopathology. The Cronbach alpha in the present study was 0.84.

Procedure and statistical analysis

The protocol was approved by the Institutional Review Board of Kaohsiung Medical University. All subjects provided written informed consent. Two research psychiatrists performed the semi-structured interviews to determine the level of insight on the SAI-E and the psychopathology on the BPRS for all subjects. Demographic characteristics, duration of illness since initial diagnosis, having psychotic features in the course of illness, and numbers of previous admission to psychiatric wards were also collected.

Data analysis was performed using SPSS 12.0 statistical software (SPSS, Chicago, IL, USA). Analysis of covariance (ancova) was used to compare the levels of insight among the subjects with SZ, BP, or DD with or without comorbid AUD, with demographic characteristics, duration of illness, numbers of previous admissions, having psychotic features, and psychopathology used as covariates. The post-hoc test with Fisher's least significant difference test (LSD) was used to further compare the level of insight among subjects with different psychiatric diagnoses. We also examined the difference in the level of insight between those with alcohol dependence and with alcohol abuse using multiple regression with regard to psychiatric diagnosis. P < 0.05 was used to indicate significance for all statistical tests.

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Patient characteristics in the SZ, BP, and DD groups with or without comorbid AUD are shown in Table 1. The mean scores on the BPRS among groups with various diagnoses ranged from 2.2 to 17.6, which indicated that most patients recruited into the present study were in a clinically stable state. The results of ancova comparing the level of insight among groups with different diagnoses, using demographic and illness characteristics as covariates, are shown in Table 2. The result indicated that the level of insight was different among groups with different diagnoses.

Table 1.  Subject characteristics
 SZ with AUDSZ without AUDBP with AUDBP without AUDDD with AUDDD without AUD
  1. AUD, alcohol use disorders; BP, bipolar I disorder; BPRS, Brief Psychiatric Rating Scale; DD, depressive disorder; SZ, schizophrenia.

Demographic characteristics
 Sex: male, n (%)46 (90.2)53 (79.1)14 (73.7)25 (62.5)53 (79.1)25 (61.0)
 Age (years) (mean ± SD)38.3 ± 7.439.6 ± 9.940.5 ± 13.240.1 ± 11.539.8 ± 10.242.7 ± 12.6
 Duration of education (years) (mean ± SD)10.3 ± 2.312.4 ± 2.810.4 ± 3.113.2 ± 2.311.3 ± 3.112.1 ± 2.9
Characteristics of illness
 Duration of illness (years) (mean ± SD)11.7 ± 6.814.9 ± 9.113.5 ± 11.511.6 ± 8.49.0 ± 8.15.2 ± 4.7
 No. previous admissions, n (%)6.7 (6.9)2.5 (2.8)3.7 (4.4)3.0 (3.5)3.6 (4.9)0.2 (0.7)
 Having psychotic features, n (%)51 (100)67 (100)7 (36.8)10 (25)23 (34.3)2 (4.9)
 Psychopathology on the BPRS (mean ± SD)17.6 ± 9.516.5 ± 10.26.4 ± 4.22.2 ± 3.19.7 ± 6.86.0 ± 4.4
Characteristics of alcohol drinking
 Duration from the first alcohol drinking (years) (mean ± SD)18.6 ± 9.219.2 ± 18.218.8 ± 12.717.6 ± 10.317.5 ± 10.818.8 ± 13.2
 Drinking alcohol once at least per month, n (%)51 (100)6 (9.0)19 (100)065 (97.0)4 (9.8)
 Drinking at least one standard drink unit of alcohol per occasion, n (%)51 (100)10 (14.9)19 (100)2 (5)67 (100)7 (17.1)
 Others have ever suggested stopping or reducing alcohol drinking, n (%)47 (92.2)6 (9.0)19 (100)1 (2.5)66 (98.5)0
 Diagnosis of alcohol dependence, n (%)30 (58.8)07 (36.8)048 (71.6)0
Table 2.  Statistical analysis of insight vs psychiatric diagnosis (ancova)
 SSd.f.Mean squareF
  • *

    P < 0.05;

  • **

    P < 0.01;

  • ***

    P < 0.001.

  • BPRS, Brief Psychiatric Rating Scale; SS, sum of square of deviations from the mean.

Diagnosis1093.6485218.7305.156***
Sex194.2751194.2754.579*
Age76.983176.9831.815
Duration of education498.6231498.62311.753**
Duration of illness0.32010.3200.008
No. previous admissions3.52813.5280.083
Psychotic features0.43510.4350.010
Psychopathology on the BPRS572.5491572.54913.496***

The means of insight in six groups using ancova adjusting for demographic and illness characteristics, and post-hoc test (LSD) between subjects with different diagnoses, are shown in Table 3. The results indicated that the DD group without comorbid AUD had a higher level of insight than did the SZ and BP groups with or without comorbid AUD. Insight in the DD group with comorbid AUD was higher than that in the SZ group with or without comorbid AUD and the BP group with comorbid AUD, but was not significantly different to that in the BP group without comorbid AUD. Insight in the BP without comorbid AUD was higher than that in the SZ group with comorbid AUD, but was not significantly different to that in the SZ group without comorbid AUD. No difference in the level of insight was found between the BP group with comorbid AUD and the SZ group with or without comorbid AUD. Although the mean insight in patients without comorbid AUD was higher than that in those with comorbid AUD (19.058 vs 16.819 in the DD group, 15.143 vs 12.974 in the BP group, and 12.947 vs 10.570 in the SZ group), the differences in insight between those with and without comorbid AUD were not large enough to reach statistical significance.

Table 3.  Means of insight on ancova adjusting for demographic and illness characteristics, and on post-hoc test (LSD)
 SZ with AUDSZ without AUDBP with AUDBP without AUDDD with AUDDD without AUD
  • *

    P < 0.05;

  • **

    P < 0.01;

  • ***

    P < 0.001.

  • AUD, alcohol use disorders; BP, bipolar I disorder; DD, depressive disorder; LSD, Fisher's least significant difference test; SZ, schizophrenia.

Mean ± SE10.570 ± 1.13312.947 ± 1.03112.974 ± 1.55315.143 ± 1.19116.819 ± 0.85219.058 ± 1.259
SZ with AUD      
SZ without AUD−2.377     
BP with AUD−2.403−0.026    
BP without AUD−4.572*−2.195−2.169   
DD with AUD−6.249***−3.872**−3.846*−1.676  
DD without AUD−8.488***−6.111**−6.084**−3.915*−2.239 

The difference in the level of insight between those with alcohol dependence and with alcohol abuse was examined on multiple regression versus psychiatric diagnosis. The results indicated that after adjusting for demographic and illness characteristics, no difference in the level of insight between those with alcohol dependence and with alcohol abuse was found within the groups of participants with SZ (β = −0.056, t = −0.334, P > 0.05), BP (β = −0.207, t = −0.830, P > 0.05), or DD (β = −0.182, t = −1.265, P > 0.05).

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

This study is one of the first to examine the effects of psychiatric diagnoses and comorbid AUD simultaneously on the level of insight. We found that no matter whether subjects had comorbid AUD or not, patients with DD had higher levels of insight than did patients with SZ, although the differences in insight between the DD group without AUD and the SZ group (−8.488 for comorbid AUD, −6.111 for non-comorbid AUD) were more significant than those between the DD group with AUD and the SZ group (−6.249 for comorbid AUD, −3.872 for non-comorbid AUD). The present study shows, again, that patients with SZ have different insight toward their illness, symptoms and necessity of treatment compared to that of DD patients.

Comorbid AUD, however, had influence on comparisons of insight between patients with DD and BP and between patients with BP and SZ. For example, although both the DD groups with and without comorbid AUD had a higher level of insight than did the BP group with comorbid AUD, only the DD group without comorbid AUD had a higher level of insight than did the BP group without comorbid AUD. Meanwhile, although the BP group without comorbid AUD had a higher level of insight than the SZ group with comorbid AUD, the level of insight in neither BP group was significantly different to that in the SZ group without comorbid AUD. This indicates that comorbid AUD may reduce the differences in insight between the DD and BP groups and between the BP and SZ groups.

The present study found that among patients with the same diagnosis, the differences in the level of insight between those with and without comorbid AUD was not significant, which was contrary to the hypothesis. Because the number of patients in each group was not large, we suggest that further studies with higher numbers of participants are needed to examine the difference in the level of insight between patients with the same psychiatric diagnosis with and without AUD.

The most unique contribution of the present study was in examining the differences in the level of insight among outpatients with different psychiatric diagnoses and in finding that comorbid AUD had an independent influence on the differences in the level of insight between patients with DD and BP and between patients with BP and SZ. This indicates that in addition to psychotic features and clinical states, comorbid AUD should be taken into consideration when comparing the level of insight among patients with different psychiatric diagnoses. It would be interesting to investigate in further studies whether the effect of comorbid AUD on insight is mediated by neurocognitive dysfunction caused by alcohol drinking.

Given that alcohol has neurotoxic effects that frequently result in significant neurocognitive deficits,19 one might have predicted that those with alcohol dependence have poorer insight than those with alcohol abuse. This hypothesis, however, was not supported by the present results. Because it was necessary to control for the effect of psychiatric diagnosis, the numbers of participants with alcohol dependence and with alcohol abuse were small; thus, further studies are needed to reach a conclusion about the association between the level of insight and severity of AUD.

Some limitations of the present study should be addressed. As previously mentioned, the number of participants in each group was not large. Recall bias is likely present given that the data were gathered through self-report. The data were provided by the participants themselves, and the authenticity with regard to the validity of some data such as the severity of alcohol drinking cannot be easily quantified. Additionally, because the present study sample consisted of outpatients who presented with less severe psychopathology as indicated by their BPRS scores, the results may not be applicable to a more severe patient population.

REFERENCES

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES