مقایسه ویژگی های بالینی در دیدگاه خوب و ضعیف اختلال وسواسی- اجباری

Beginning with DSM-III-R, the condition of an intact insight towards obsessive–compulsive symptoms, which was essential for the classical definition of obsessive–compulsive neurosis, has been removed, permitting inclusion of cases with poor insight. A total of 94 cases who met DSM-III-R criteria for obsessive–compulsive disorder were included in this study. The Structured Clinical Interview for DSM-III-R (SCID-P), Yale–Brown Obsessive Compulsive Scale (Y-BOCS), Hamilton Rating Scale for Depression (HRSD), Hamilton Rating Scale for Anxiety (HRSA), and State–Trait Anxiety Inventory (STAI) were administered to each patient. Two subgroups determined by DSM-IV item “poor insight” were compared for demographic variables and the scores obtained on the scales. Scores on the Y-BOCS, HRSA, HRSD and STAI-state were significantly higher in the poor insight group. Current and past major depression were also more frequent. Among personality disorders (PDs), avoidant PD was more common in the good insight group and borderline and narcissistic PDs were more common in the poor insight group. HRSA, HRSD, and STAI-state scores had weak to moderate but significant correlations with insight as defined by the item 11 of Y-BOCS. Findings are discussed in view of previous reports.

The problem of insight has been a key issue in the series of debates surrounding the definition of obsessive–compulsive disorder (OCD). First of all, obsessions and compulsions also occur in psychoses, and this calls for a clear-cut boundary between neurotic and psychotic obsessions and compulsions. The traditional gold standard in making the distinction has been presence of insight in the former (Reed, 1985 and Snaith, 1981). Secondly, as many authors addressing OCD have noted, presence and degree of insight may change considerably from one patient to another and over the course of the disorder (Freud, 1966; Insel & Akiskal, 1986; Kozak & Foa, 1993; Lelliott, Noshirvani, Basoglu, Marks, & Monteiro, 1988; Robinson, Winnik, & Weiss, 1976; Solyom, DiNicola, Phil, Sookman, & Luchins, 1985), making such a distinction difficult at times, if not altogether impossible. And thirdly, it has been noted that patients with poor insight also respond to treatment poorly (Basoglu, Lax, Kasvikis, & Marks, 1988; Eisen & Rasmussen, 1993; Foa, 1979; Insel & Akiskal, 1986; Jenike, Baer, Minichiello, Schwartz, & Carey, 1986; Solyom et al., 1985), raising the question of whether there is a subgroup of “atypical” OCD patients whose symptoms are more severe. In case such a group exists, would it not be more appropriate to classify these patients under a separate entity or under psychoses? Hoch and Polatin described a case with “pseudoneurotic schizophrenia” in as early as 1949 (cited in Spitzer, Skodol, Gibbon, & Williams, 1981). Strauss (1948), Weiss, Robinson, and Winnik (1969) and also Solyom et al. (1985) offered “obsessive psychosis,” Insel and Akiskal (1986) offered “OCD with psychotic features,” and Rasmussen and Tsuang (1986) offered “chronic deteriorative OCD.” However, subpopulations defined by those terms are not necessarily identical. Appropriateness of classifying some “atypical” cases with poor insight under the classical entity of OCD has been questioned by many authors in recent years. The first one to note that obsessive patients do not necessarily have insight into the senselessness of their beliefs was Lewis (1936) who, as Jakes (1996) pointed out, also employed the term “resistance” in a special way to broaden the definition of OCD. However, with the attempts to use uniform criteria in classification getting stronger in DSM-III (American Psychiatric Association, 1980), the definition became somewhat stricter. Still, it may be interesting to note that in the historical cases section of the official casebook for DSM-III, Spitzer et al. (1981) re-diagnosed a case with probably negative insight as OCD, arguing against the original diagnosis of “pseudoneurotic schizophrenia” offered by Hoch and Polatin. The tendency to broaden the definition gained ground in time and resulted in a loosening of the criterion “insight into senselessness” in DSM-III-R (American Psychiatric Association, 1987), which admitted that this “ … may no longer be true for people whose obsessions have evolved into overvalued ideas.” DSM-IV went even further by stating that obsessions or compulsions must be recognized as excessive or unreasonable not necessarily earlier, but “at some point during the course of the disease.” DSM-IV also stressed that the specifier “with poor insight” was added “in recognition that insight … occurs on a continuum” (American Psychiatric Association, 1994), in line with Kozak and Foa (1993) who had pointed out the difficulty of drawing lines between obsession and overvalued idea and overvalued idea and delusion, which can be distinguished only by the degree of insight present. In this study, we investigated 94 OCD cases with varying degrees of insight and tried to find out whether these patients constitute a homogenous subgroup. We also tried to determine the other features that may help to characterize such a subgroup.

Of the 94 cases with OCD, 29 (31%) had poor insight defined according to DSM-IV criteria. There was no statistically significant difference between the groups for gender, marital status or education (Table 1). The poor insight group tended to include more males. No difference between the two groups was observed for the mean duration of the disease, age at onset, and age at assessment. When the dichotomous variable of insight was replaced by the more sophisticated Y-BOCS item 11, none of the above-mentioned variables displayed any correlation with it. Table 1. Socio-demographic characteristics of the good and poor insight subgroups Good insight (N = 65) Poor insight (N = 29) χ2 N (%) N (%) Gender Female 41 (63.1) 13 (44.8) 2.73* Marital status Unmarried 40 (61.5) 16 (55.2) 3.86 Married 23 (28.1) 9 (31.0) Divorced/widow 2 (2.2) 4 (13.8) Education 5 years 15 (23.1) 8 (27.6) .55 8 years 7 (10.8) 2 (6.9) 11 years or more 43 (66.2) 19 (65.5) Note: Good/poor insight was defined according to DSM-IV specifier. * P<.1. Table options Obsession, compulsion, and total scores as determined by the Y-BOCS were significantly higher for the poor insight group; as were the STAI-state, HRSD and HRSA scores (Table 2). An overall MANOVA conducted on HRSD, HRSA, STAI-state, and STAI-trait, Y-BOCS obsession, Y-BOCS compulsion and total scores indicated no significant gender differences, F(6,85)=.95, P>.05. Table 2. Age, age at onset, duration and mean scores on measures of anxiety, depression and obsessive–compulsive symptomatology Good insight (N = 65) Poor insight (N = 29) t Mean (S.D.) Mean (S.D.) Age at assessment 28.2 (9.5) 28.6 (10.5) −.16 Age at onset 21.2 (8.3) 20.9 (9.6) −.14 Duration (month) 83.4 (65.6) 91.7 (83.2) −.48 HRSA 18.2 (9.9) 23.8 (11.0) −2.35** HRSD 12.2 (7.9) 17.0 (8.8) −2.53** STAI-S 45.1 (11.1) 51.6 (12.5) −2.42** STAI-T 50.2 (11.4) 55.0 (12.4) −1.79* Y-BOCS obsession 12.5 (3.8) 14.5 (3.5) −2.53** Y-BOCS compulsion 11.9 (4.0) 14.9 (3.6) −3.58*** Y-BOCS total 24.4 (7.0) 29.4 (5.8) −3.65*** Note: Good/poor insight was defined according to DSM-IV specifier. HRSA, Hamilton Rating Scale for Anxiety; HRSD, Hamilton Rating Scale for Depression; STAI-S, State–Trait Anxiety Inventory, state; STAI-T, State–Trait Anxiety Inventory, trait; Y-BOCS, Yale–Brown Obsessive Compulsive Scale. * P<.1. ** P<.05. *** P<.01. Table options There was no significant difference between the subgroups in terms of types of obsessions and compulsions, except for a statistically significant excess of “miscellaneous” compulsions in the poor insight group. To further test the hypothesis that insight occurs on a continuum, with varying degrees of impairment corresponding to various degrees of syndrome severity, we examined the relationship between scores obtained in item 11 of Y-BOCS (insight score 0–4), and scores on measures of anxiety, depression and obsessive–compulsive symptomatology. There were significant positive correlations between the Y-BOCS item 11 scores and Y-BOCS obsessive, compulsive, and total scores (r=.30, P<.01, r=.40, P<.001, and r=.38, P<.001, respectively). The Y-BOCS item 11 scores were significantly correlated with the HRSD, HRSA and STAI-state scores (r=.27, P<.01; r=.21, P<.05; and r=.24, P<.05, respectively). STAI-trait scores also tended to correlate with the Y-BOCS item 11 scores; but this trend did not reach the level of statistical significance (r=0.20, P<.1). We tested the probable effects of mood and anxiety measures on the relationship between insight and Y-BOCS scores with a stepwise linear regression model. Y-BOCS total scores emerged as the single most important predictor of insight as defined by Y-BOCS item 11 (r2=.17, S.E.=1.37, F=18.52, β=.41, P<.001). The model excluded HRSD, HRSA, STAI-state, and STAI-trait scores, as well as Y-BOCS obsession and Y-BOCS compulsion subscores. Major depression, both current and past, was the only Axis I disorder seen more frequently in the poor insight group (Table 3). The Mann–Whitney test with Wilcoxon rank sum showed that scores obtained in Y-BOCS-11 were consistently higher in currently depressed patients (U=733, W=2624, z=2.25, P<.05). A correlation analysis verified that it was more likely to find current major depression as Y-BOCS-11 scores went higher (Spearman’s ρ=.23, P<.05). Table 3. Frequency and percentage of additional Axis I diagnoses Poor insight (N = 29) Good insight (N = 65) χ2 N (%) N (%) Panic disorder 5 (7.7) 3 (10.3) .18 Generalized anxiety disorder 4 (6.2) 4 (13.8) 1.50 Social phobia 7 (10.8) 4 (13.8) .18 Simple phobia 9 (13.8) 9 (31.0) 3.68 Major depression (current) 18 (27.7) 15 (51.7) 5.08* Major depression (past) 18 (27.7) 15 (51.7) 5.08* Note: Good/poor insight was defined according to DSM-IV specifier. Axis I disorders occurring in less than four patients were excluded from the table. * P<.05. Table options The SCID-II was administered to a total of 50 patients. Avoidant personality disorder was more frequent in the good insight group while narcissistic and borderline PDs were more frequent in the poor insight group. There was no patient with schizotypal personality disorder in the poor insight group while there was one in good insight group. The poor insight group had also significantly higher rates of cluster B PDs than the good insight group (P<.01) ( Table 4). Analysis of the relationships between DSM-defined personality disorder clusters and Y-BOCS-11 scores showed that patients with cluster B (U=125.5, W=828.5, z=2.71, P<.01) but not cluster A or C disorders tended to have higher scores. A correlation analysis verified the relationship (Spearman’s ρ=.39, P<.01). Table 4. Frequency and percentage of personality disorders Good insight (N = 40) Poor insight (N = 10) χ2 N (%) N (%) Obsessive–compulsive 17 (42.5) 5 (50.0) .18 Avoidant 15 (37.5) 0 (0) 5.36* Dependent 3 (7.5) 1 (10.0) .07 Passive–aggressive 4 (10.0) 1 (10.0) 0 Histrionic 4 (10.0) 2 (20.0) .76 Borderline 2 (5.0) 4 (40.0) 9.28* Narcissistic 1 (2.5) 3 (30.0) 8.22* Paranoid 5 (12.5) 1 (10.0) .05 Schizoid 4 (10.0) 0 (0.0) 1.09 Cluster A 8 (20.0) 1 (10.0) .54 Cluster B 6 (15.0) 7 (70.0) 12.58** Cluster C 25 (62.5) 6 (60.0) .02 Note: Good/poor insight was defined according to DSM-IV specifier. Schizotypal personality disorder occurring in only one patient in good insight group was excluded from the table. * P<.05. ** P<.01.