SCHRES-06208; No of Pages 7 Schizophrenia Research xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
The diagnostic criteria and structure of catatonia Jo Ellen Wilson a,⁎,1, Kathy Niu a,1, Stephen E. Nicolson a, Stephen Z. Levine b, Stephan Heckers a a b
Department of Psychiatry, Vanderbilt University, Nashville, TN, USA Department of Community Mental Health, University of Haifa, Israel
a r t i c l e
i n f o
Article history: Received 31 January 2014 Received in revised form 18 December 2014 Accepted 22 December 2014 Available online xxxx Keywords: Catatonia DSM-IV DSM-5 Diagnosis Principal component analysis Item response theory
a b s t r a c t Background: The classification of catatonia has fluctuated and underwent recent changes in DSM-5. The current study examines the prevalence of catatonia signs, estimates the utility of diagnostic features, identifies core catatonia signs, and explores their underlying structure. Method: We screened 339 acutely ill medical and psychiatric patients with the Bush Francis Catatonia Rating Scale (BFCRS). We examined prevalence and severity of catatonia signs and compared BFCRS, DSM-IV and DSM-5 diagnoses. We used principal component analysis (PCA) to examine the factorial validity of catatonia and item response theory (IRT) to estimate each sign's utility and reliability. Results: Out of the 339 patients, 300 were diagnosed with catatonia using the BFCRS and 232 catatonia diagnoses were validated by the treating provider based on selection for treatment with benzodiazepines or electroconvulsive therapy. Of the 232 validated catatonia cases, 211 (91%) met DSM-IV criteria but only 170 (73%) met DSM-5 criteria for catatonia. Staring was the most prevalent catatonia sign. PCA identified three components, interpretable as “Increased, Abnormal and Decreased Psychomotor Activity,” although 63% of the variance was unexplained. IRT showed that Excitement, Waxy Flexibility and Immobility/Stupor were the best indicators of each factor. The BFCRS had many redundant items and as a whole had low reliability at low severity of catatonia, but good reliability at moderate–high severity of catatonia. Conclusions: The structure of catatonia remains to be discovered. Published by Elsevier B.V.
1. Introduction In 1874, Karl Kahlbaum described catatonia in patients who suffered from severe psychotic, mood and medical conditions (Kahlbaum, 1874). Kraepelin and Bleuler, however, redefined catatonia as a subtype of dementia praecox (1896) (Kraepelin, 1919) and schizophrenia (1911) (Bleuler, 2010). Clinicians followed their redefinition throughout most of the 20th century, with a few notable exceptions (Morrison, 1973, 1974a,b, 1975; Abrams and Taylor, 1976, 1977; Taylor and Abrams, 1977; Abrams et al., 1979). The diagnostic categories for catatonia were broadened in DSM-III-R (catatonia was added as a specifier for mood disorders) and DSM-IV (catatonia resulting from a general medical condition was added) (AP, 2000). DSM-5 introduced several changes in the classification of catatonia, intended to improve recognition, treatment and research (Taylor and Fink, 2003; Fink et al.,
⁎ Corresponding author at: Vanderbilt Psychiatric Hospital, 1601 23rd Ave. South, Nashville, TN 37212, USA. Tel.: +1 615 936 3555. E-mail addresses: [email protected] (J.E. Wilson), [email protected] (K. Niu), [email protected] (S.E. Nicolson), [email protected] (S.Z. Levine), [email protected] (S. Heckers). 1 Dr. Wilson and Dr. Niu are joint first authors.
2010; Francis et al., 2010; Heckers et al., 2010; Rosebush and Mazurek, 2010; Braff et al., 2013; Tandon et al., 2013). Despite the changes in nosology, it is unclear which and how many signs define a case of catatonia. More than 40 clinical signs of catatonia have been described and at least seven scales for the assessment of catatonia have been proposed (Taylor and Fink, 2003). The preferred rating scale for routine catatonia assessment is the 23-item BushFrancis Catatonia Rating Scale (BFCRS) because of reports of its validity and reliability, ease of administration and since cases can be identified based on the presence of 2 or more items of the first 14 items (Sienaert et al., 2011). In contrast, the DSM-5 criteria require a minimum of 3 out of 12 signs for the diagnosis of catatonia. How these differences in the definition of caseness affect the recognition of catatonia is currently unknown. Previous studies have identified up to six dimensions of catatonia (Abrams et al., 1979; McKenna et al., 1991; Starkstein et al., 1996; Oulis et al., 1997; Northoff et al., 1999; Peralta et al., 1999, 2010; Peralta and Cuesta, 2001; Kruger et al., 2003; Ungvari et al., 2007). Except for one study (Oulis et al., 1997), the factors stupor and excitement are a common finding, although the naming and exact grouping of signs differ. Here we report the analysis of a diverse clinical sample of acutely ill medical and psychiatric patients with catatonia. We compare diagnosis
http://dx.doi.org/10.1016/j.schres.2014.12.036 0920-9964/Published by Elsevier B.V.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
2
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
rates based on BFCRS, DSM-IV and DSM-5, study the factorial structure of catatonia and estimate the reliability of the catatonia signs in the BFCRS. 2. Methods 2.1. Registry Patients were identified using a retrospective chart review. We included all patients with a completed Bush Francis Catatonia Rating Scale (BFCRS) assessment in the electronic medical record system at Vanderbilt University Medical Center between 06/18/2009 and 06/16/ 2012. All catatonia assessments were performed by psychiatrists who had experience in assessing catatonia and were trained in the use of the BFCRS. We identified 339 patients with a BFCRS assessment performed in one of four settings: general medical hospital, emergency room, psychiatric hospital or outpatient psychiatry clinic. If a patient was represented with more than one rating scale, only the first assessment performed by the most senior evaluator was included for analysis. All 339 patients were screened for catatonia with the 14-item Bush-Francis Catatonia Screening Instrument (BFCSI). 300 patients (88.5%) scored for 2 or more items and were assessed with the full 23-item BFCRS (see Supplemental Fig. 1). Careful case-by-case review of all available medical records was undertaken to determine the underlying medical and psychiatric diagnoses (see Supplemental Fig. 2). Additionally, response to treatment with benzodiazepines and ECT was scored as “substantially improved”, “partially improved” or “not improved”, as documented in the notes by the treating physician. 2.2. BFCSI and BFCRS The Bush Francis Catatonia Rating Scale (BFCRS) is the most widely used catatonia rating scale, given its ease of use and high inter-rater reliability and validity (Sienaert et al., 2011). The scale consists of 23 signs, with the first 14 items constituting the Bush Francis Catatonia Screening Instrument (BFCSI). The definition of catatonia proposed by Bush et al. (1996a) is the presence of two or more signs on the BFCSI, regardless of severity (Bush et al., 1996a,b). Sixteen items are rated on a severity scale 0 (absent) through 3 (e.g. constant), while the other six items are rated 0 (absent) and 3 (present). The total BFCRS score is the sum of the responses to all 23 items. 2.3. Catatonia diagnoses The presence of 2 or more BFCSI signs on examination meets the standard for a diagnosis of catatonia (Bush et al., 1996a,b). Within the 300 patients who met the BFCSI diagnosis of catatonia, we indentified a subset of 232 patients meeting the following 3 criteria: 1) at least 2 items on BFCSI, 2) clinical diagnosis of catatonia by treating provider and 3) selection for treatment with a benzodiazepine or ECT. We will focus our report on this subset of 232 catatonia patients validated by treatment. To determine caseness for catatonia we calculated the number of positive items on the BFCSI for each subject and identified which signs, if removed, lowered the sum below 2. DSM-IV and DSM-5 diagnoses of catatonia were established by applying the respective diagnostic criteria A (APA, 2013) to all cases with 2 or more BFCSI signs. To ascertain sensitivity of diagnostic criteria we examined the severity and prevalence of catatonic signs, catatonia caseness, the effect of treatment and the potential confounding effect of psychopharmacological treatment history. 2.4. Psychometric analysis Data analysis proceeded in three steps. Step one consisted of computing item (i.e., sign) frequencies. In step two, principal components analysis (PCA) (Kruskal and Wish, 1978; Gorsuch, 1993;
Tabachnick and Fidell, 2001) was computed to identify the number of dimensions that underlie the BFCRS. To identify the most parsimonious number of dimensions, a scree plot was generated, and results from 1000 simulations of equal data size and symptoms were superimposed on the graph. This was supplemented by considerations for reliability and interpretability. Promax rotation allows for correlation between dimensions and was used to rotate the structural coefficients to interpret the dimensions (Gorsuch, 1993). The cut-off value for item loadings was 0.40. In the third step, item response theory (IRT) analysis was computed with Latent Trait Models (LTM) (Rizopoulos, 2006) (Levine et al., 2011). IRT quantifies the disparity between clinical ratings and the expression of the underlying (latent) catatonia phenotype.
3. Results 3.1. Demographic and clinical characteristics Patients were assessed with the BFCSI in a psychiatric hospital (n = 218, 64.3%), a tertiary medical center (n = 66, 19.5%), an emergency room (n = 44, 13.0%) and a psychiatry outpatient clinic (n = 11, 3.2%) (Supplemental Table 1).
3.2. Severity and prevalence of catatonia signs In the group of 339 patients screened, we observed a mean of 5.27 (2.39 SD) signs using the BFCSI and 7.58 (3.51 SD) signs using the full BFCRS. Staring was the most prevalent sign (observed in 70.5% of all patients) and combativeness was the least frequently reported sign (observed in 10.9% of all patients) (Fig. 1A). 20 patients (8.6% of all patients) showed only 2 signs, i.e., the minimum number of signs for caseness (see Fig. 1B). For the 232 validated cases of catatonia, the mean BFCSI and BFCRS scores were 5.60 ± 2.34 and 8.08 ± 3.45 respectively.
3.3. Catatonia caseness We assessed how often cases would fail to meet the criteria for catatonia if an item was removed from the criterion set for the BFCSI, DSM-IV and DSM-5 (Fig. 1C). Staring had the greatest impact on caseness, followed by immobility/stupor, excitement, rigidity and mutism. Grimacing, echopraxia/echolalia and waxy flexibility had no effect on caseness. Of the 232 patients with validated catatonia, 211 (90.95%) met DSM-IV criteria but only 170 (73.28%) met DSM-5 criteria for catatonia.
3.4. Effect of treatment with benzodiazepines and ECT Of the 232 cases selected for treatment with benzodiazepines or ECT, 7 (3%) patients refused the recommended treatment. The average effective dose of lorazepam required for treatment response was 5.81 mg per day (SD 4.7). 6 patients (2.6%) were treated with clonazepam, requiring an average dose of 2.8 mg per day until clinical improvement. Of the 225 patients treated with a benzodiazepine, 191 (84.9%) substantially improved, 25 (11.1%) partially improved, 9 (3.1%) did not respond to treatment with a benzodiazepine. 45 (19.4%) patients required treatment with ECT, generally due to the presence of severe catatonia (autonomic abnormalities, refusal to eat or drink), or incomplete response to benzodiazepines. 42 patients (93.3% of those who received ECT) substantially improved, while 3 patients (6.7%) partially improved. The average number of ECT sessions required during induction for treatment remission was 8.4 (SD 5.6). 1 patient was already undergoing maintenance ECT at the time catatonia was assessed.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
Staring Immobility/Stupor Musm Withdrawal Posturing/Catalepsy Negavism Automac Obedience Rigidity Autonomic Abnormality Ambitendency Mitgehen Stereotypy Impulsivity Perseveraon Excitement Grimacing Verbigeraon Waxy Flexibility Gegenhalten Echopraxia/Echolalia Mannerisms Grasp reflex Combaveness
80% 65% 56% 49% 45% 44% 41% 38% 32% 31% 28% 27% 27% 26% 24% 24% 21% 20% 20% 18% 15% 14% 12%
B
C 18%
Waxy Flexibility Echopraxia/echolalia Grimacing Verbigeraon Withdrawal Negavism Stereotypy Mannerisms Posturing/catalepsy Rigidity Musm Excitement Imobility/Stupor Staring
16% 14%
BFCSI BFCRS
Percentage of Paents
A
3
12% 10% 8% 6% 4% 2% 0% 0
1
2
3
Frequency of catatonia signs
4
5
6
7
8
9 10 11 12 13 14
Number of catatonia signs
0
1
2
3
4
5
Caseness of catatonia signs (%)
Fig. 1. Prevalence, severity and caseness of catatonia signs. A. Prevalence of catatonia signs in 300 patients with a score of ≥2 on the BFCSI. BFCSI items are shown in black, remaining BFCRS items are shown in gray. B. Severity of catatonia, as measured by the number of catatonia signs in patients screened with the BFCSI. C. Effect of caseness after removal of individual BFCSI items.
3.5. Effect of treatment with antipsychotic medication 103 (44.4%) of the 232 patients with validated catatonia were being treated with either a typical or atypical antipsychotic medication at the time of the assessment (Table 1). Treated and untreated patients did not differ in the mean number of catatonia signs, but the prevalence of various items differed (see Supplemental Fig. 3). Patients treated with an antipsychotic drug showed significantly (p b 0.05) less mutism, staring, grimacing, withdrawal and significantly (p b 0.05) more automatic obedience, mitgehen and impulsivity. 3.6. Prevalence of catatonia signs Table 2 shows the prevalence of the catatonia signs in the 232 patients meeting BFCRS criteria. Only four symptoms exceeded a frequency of 50%: Staring (N = 189, 81.5%), Immobility/Stupor (N = 162, 69.8%), Mutism (N = 144, 62.1%) and Withdrawal (N = 122, 52.6%). For all items, the most severe rating of “3” was the least selected rating option. For Stereotypy, none responded with a “3.” The four symptoms rated most frequently as absent were Combativeness (N = 206, 88.79%), Grasp Reflex (N = 200, 86.21%), Mannerisms (N = 194, 83.62%) and Stereotypy (N = 186, 80.17%). 3.7. Psychometric analysis: principle component analysis (PCA) Here we report the psychometric analysis (PCA and IRT) of the 232 validated catatonia cases.
A three-factor solution explained 37% of the total variance (Table 3 and Supplemental Fig. 4). Factor 1 consisted of Excitement, Impulsivity, Combativeness, Verbigeration, Mannerisms, Grimacing, and Stereotypy, and was hence labeled “Increased Psychomotor Activity.” Factor 2 consisted of Waxy Flexibility, Automatic Obedience, Mitgehen, Ambitendency, Grasp Reflex, Gegenhalten, and Posturing/Catalepsy, and was hence labeled “Abnormal Psychomotor Activity”. Factor 3 consisted of Immobility/Stupor, Mutism, Withdrawal, Staring, and Negativism, and was hence labeled “Decreased Psychomotor Activity”. Perseveration, Echopraxia/Echolalia, Autonomic Abnormality, and Rigidity did not load on a factor. The three factors showed moderate correlations at best, the highest between factors 3 (Decreased Psychomotor Activity) and 2 (Abnormal Psychomotor Activity) (correlation 0.241). PCA of the 300 cases meeting BFCRS criteria for catatonia also revealed 3 factors, but the cumulative variance ranking was slightly different with Increased Psychomotor Activity (15.09%), Decreased Psychomotor Activity (28.75%) and Abnormal Psychomotor Activity (35.96%). 3.8. Psychometric analysis: item response theory (IRT) Item response theory (IRT) was computed for each of the BFCRS factors to test the strength of each item's relationship between ratings and the latent phenotype catatonia (Supplemental Table 1). To quantify this relationship, we defined a discrimination parameter value (α) of 1 as a data-driven cut-off value. The Factor 1 items with
Table 1 Demographic and clinical characteristics.
Demographics Age Gender (F/M) Race (Caucasian/AA/other) Education Clinical characteristics Psychiatric hospital Medical center Emergency room Psychiatry Clinic Number of BFCSI items present Number of BFCRS items present Chlorpromazine equivalent; % of subjects treated
Patients with catatonia, validated by treatment n = 232
Patients with catatonia, not validated by treatment n = 68
Patients without catatonia n = 39
38.37 ± 18.68 128/104 151/65/16 12.16 ± 2.98
36.10 ± 18.12 34/34 39/22/7 11.31 ±2.79
35.54 ± 18.89 16/23 20/16/3 11.92 ± 2.53
153 (65.95) 44 (18.97) 27 (11.64) 8 (3.45) 5.6 ± 2.34 8.08 ± 3.45 173.26 ± 376.08; (44.40)
43 (63.24) 13 (19.12) 11 (16.18) 1 (1.47) 4.12 ±2.24 5.97 ±3.26 90.66 ±191.06; (33.82)
22 (56.41) 9 (23.08) 6 (15.38) 2 (5.13) 0.74 ± 0.64 2.28 ± 2.80 184.34 ± 277.4; (51.28)
Numbers are mean ± SD. Percentage of sample in parentheses.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
4
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
Table 2 BFCRS item responses in 232 catatonia patients. Response
1. Excitement 2. Immobility/Stupor 3. Mutism 4. Staring 5. Posturing/Catalepsy 6. Echopraxia/Echolalia 7. Stereotypy 8. Grimacing 9. Mannerisms 10. Verbigeration 11. Rigidity 12. Negativism 13. Waxy Flexibility 14. Withdrawal 15. Impulsivity 16. Automatic Obedience 17. Mitgehen 18. Gegenhalten 19. Ambitendency 20. Grasp Reflex 21. Perseveration 22. Combativeness 23. Autonomic Abnormality
0 e.g. “absent”
1 e.g. “occasional”
2 e.g. “frequent”
3 e.g. “constant”
N
(%)
N
N
N
179 71 90 41 120 175 186 162 194 179 142 127 183 110 168 137
(77.16) 37 (15.95) 14 (6.03) 2 (.86) (30.60) 127 (54.74) 33 (14.22) 1 (.43) (38.79) 52 (22.41) 66 (28.45) 24 (10.34) (17.67) 91 (39.22) 88 (37.93) 12 (5.17) (51.72) 54 (23.28) 46 (19.83) 12 (5.17) (75.43) 41 (17.67) 12 (5.17) 4 (1.72) (80.17) 28 (12.07) 18 (7.76) 0 (.00) (69.83) 31 (13.36) 32 (13.79) 7 (3.02) (83.62) 26 (11.21) 9 (3.88) 3 (1.29) (77.16) 24 (10.34) 24 (10.34) 5 (2.16) (61.21) 55 (23.71) 30 (12.93) 5 (2.16) (54.74) 60 (25.86) 32 (13.79) 13 (5.60) (78.88) 0 (.00) 0 (.00) 49 (21.12) (47.41) 41 (17.67) 69 (29.74) 12 (5.17) (72.41) 39 (16.81) 23 (9.91) 2 (.86) (59.05) 30 (12.93) 36 (15.52) 29 (12.50)
158 180 150 200 165 206 156
(68.10) (77.59) (64.66) (86.21) (71.12) (88.79) (67.24)
(%)
(%)
0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 17 (7.33) 7 (3.02) 45 (19.40) 25 (10.78)
74 52 82 32 67 2 6
(%)
(31.90) (22.41) (35.34) (13.79) (28.88) (.86) (2.59)
an α exceeding 2 were Excitement, Impulsivity, Combativeness, Mannerisms, indicating that these were the most representative items of the “Increased Psychomotor Activity” dimension. All other loaded items — Verbigeration, Grimacing, and Stereotypy had good discrimination with α N 1 as well. Factor 2 had Waxy Flexibility with α N 2; Gegenhalten, Grasp Reflex, Mitgehen with α N 1, and Ambitendency and Posturing/Cataplexy close by. Automatic Table 3 PCA factors, loading values, and factor correlation matrix. Component
Excitement Impulsivity Combativeness Verbigeration Mannerisms Grimacing Stereotypy Perseveration Echopraxia/Echolalia Waxy Flexibility Automatic Obedience Mitgehen Ambitendency Grasp Reflex Gegenhalten Posturing/Catalepsy Autonomic Abnormality Immobility/Stupor Mutism Withdrawal Staring Negativism Rigidity Cumulative % of variance Eigenvalues Reliability Kaiser–Meyer–Olkin measure
1
2
3
0.8 0.77 0.74 0.61 0.6 0.5 0.41 0.35 0.26 −0.03 −0.08 −0.03 −0.16 0.24 0.09 −0.09 0.04 −0.33 −0.02 0.1 −0.17 0.17 0.26 16.11% 3.71 0.77 0.73
−0.13 −0.12 −0.19 0.08 −0.05 0.19 0.32 0.2 0.19 0.65 0.55 0.54 0.53 0.47 0.46 0.46 0.16 −0.04 −0.06 −0.22 0.26 0.15 0.35 29.53% 3.09 0.61
−0.13 0.1 0.24 0.05 −0.15 0 −0.16 −0.19 0.21 0.02 −0.19 0.02 −0.16 0.02 0.11 0.36 0.08 0.71 0.71 0.71 0.45 0.43 0.35 36.99% 1.72 0.64
Numbers in bold represent BFCRS items that loaded onto the respective catatonia component.
Obedience had the least strong relationship to the “Abnormal Psychomotor Activity” dimension. Factor 3 had Immobility/Stupor with α N 2; Mutism, Withdrawal, and Staring with α N1. However, Negativism had the least strong relationship to the “Decreased Psychomotor Activity” dimension. Threshold parameters (βs) indicate the severity of catatonia at which two rating response categories are equally likely to be selected. For example, β1 represents equal likelihood of selecting response 1 “absent” and response 2 “occasional.” Of note, only items in Factor 3 had a β b − 1, allowing for more information at low levels of catatonia. If a threshold value exceeds 1.96, it suggests ratings provide good information. The Item Response Category Characteristic Curves (ICC) graphically represents a combination of the item discrimination and difficulty, where the probability of a certain response is graphed against the severity of catatonia (left column of Fig. 2 and Supplemental Fig. 5). Examination of the ICCs shows that Verbigeration, Grimacing, Automatic Obedience, Withdrawal have substantial overlap between rating options “1” and “2,” suggesting that these two rating options could be collapsed into one without substantial data loss. Notably, Negativism shows poor separation of response choices, suggesting that the choices between mild, moderate, and severe resistance or contrary behavior is not informative to distinguish the severity of Factor 3. The information curves show how much information or reliability an item offers relative to the severity of catatonia (right column of Fig. 2). Most items peak around the same range of catatonia severity, which suggests redundancy. The individual item information curves combine to form the test information curve for that set of items. The test information curve shows that the set of items for Factor 1 is reliable for mostly mid-high level of “Increased Psychomotor Activity” catatonia, Factor 2 items are reliable at mid-high levels of “Abnormal Psychomotor Activity” catatonia, and Factor 3 items are reliable at low, medium, and high levels of “Decreased Psychomotor Activity” catatonia. No set of items are reliable at very low levels of catatonia. 4. Discussion We assessed the prevalence of catatonia signs in a large clinical sample, tested the effect of different criteria sets on categorical diagnoses and explored its core factor structure. Our results suggest that the DSM-5 criteria are more restrictive than previous criteria sets, excluding more than a quarter of patients meeting BFCRS criteria from the diagnosis. Principal component analysis identified three factors labeled “Increased Psychomotor Activity”, “Abnormal Psychomotor Activity”, and “Decreased Psychomotor Activity” that cumulatively accounted for 37% of total variance. Item response theory identified that the BFCRS lacks reliability at low levels of catatonia. 4.1. Effect of diagnostic criteria on recognition of catatonia Staring is recognized as one of the most prevalent catatonia signs, with estimates ranging from 25% to 92% (18, 47–57). This was confirmed in our sample, where staring was the most prevalent catatonia sign. Yet, staring is not included in any version of the DSM. We recommend that staring be included in the diagnostic criteria for catatonia, since it can be a crucial diagnostic sign (in 4% of our patients meeting BFCSI criteria). Rigidity is recognized by most catatonia rating scales (Sienaert et al., 2011) and in earlier versions of the DSM, but was removed from the DSM-5 criteria. This is supported by our analysis, since removal of rigidity affected caseness in less than 2% of cases. 44.4% of our patients were treated with antipsychotic medication at the time of the assessment. While the severity of catatonia did not differ between treated and untreated patients, some signs were more or less prevalent in the treated patients. Antipsychotic medication may induce an extrapyramidal syndrome akin to catatonia, which may exist on a
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
Item Response Category Characterisc Curves
Item Informaon Curves
1.5
excitement impulsivity combativeness mannerisms verbigeration grimacing stereotypy
0.0
0.0
0.5
1.0
0 1 2 3
Informaon
0.6
Rating
0.4
2.0
0.8
2.5
1.0
Informaon Test Curves: Factor 1
0.2
Probability
Increased Psychomotor Acvity
5
−4
−2
0
2
−4
4
−2
Excitement
0
Catatonia
2
4
1.0
waxyflexibility automaticobedience mitgehen ambitendency graspreflex gegenhalten posturingcatalepsy
0.5
Informaon
0 1 2 3
0.4
0.6
Rating
0.0
0.0
0.2
Probability
Decreased Psychomotor Acvity
0.8
1.5
1.0
Informaon Test Curves: Factor 2
−4 −4
−2
0
Musm
2
−2
4
0
2
4
Catatonia
2.0
mutism immobilitystupor withdrawal negativism staring
0.5
1.0
1.5
Informaon
0.8
0 3
0.4
0.6
Rating
0.0
0.0
0.2
Probability
Abnormal Psychomotor Acvity
2.5
1.0
3.0
Informaon Test Curves: Factor 3
−4
−2
0
2
Waxy Flexibility
4
−4
−2
0
2
4
Catatonia
Fig. 2. Item response category characteristic curves and information curves. Item response category characteristic curves (left column) are a probabilistic index of the relationship between ratings for an item (i.e., catatonia sign) and level on the latent construct (i.e., catatonia). For example, the curve for the sign “Excitement” depicts the distinction between “absent” that is likely at low symptom severity and “two” that is likely at higher symptom severity. Item information curves (right column) index items over latent severity to assess reliability at different severity levels. For example, catatonia Factor 1 (Increased Psychomotor Activity) assesses little catatonia when the item is low (x-axis values −4 to 0), more at moderate levels (x axis values 0 to 2), and little at high levels (x axis values 2 to 4).
spectrum between simple non-malignant catatonia and malignant catatonia or NMS (Caroff et al., 2002; Vesperini et al., 2010). Although the exact incidence of drug-induced catatonia is unknown, certain catatonia signs are associated with antipsychotic medication, including akinesia/stupor, mutism, catalepsy and waxy flexibility (Lopez-Canino
and F.A., 2004). Due to the retrospective design, the authors can only speculate about the effect of neuroleptic treatment. We assume that providers were more apt to reduce antipsychotic drug exposure in those with classic catatonic features (i.e., the Kahlbaum subtype of catatonia) (Fink, 2006), thus accounting for a significantly lower
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
6
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
prevalence of mutism, staring, grimacing and withdrawal in the neuroleptic treated group. Additionally, we speculate that medical providers might have been inclined to provide a neuroleptic to an agitated patient either due to concern for violence or because of lack of recognition of the catatonic syndrome, therefore accounting for a higher prevalence of automatic obedience, mitgehen and impulsivity in the neuroleptic treated group. 4.2. Psychometric properties of BFCRS Only four items (staring, immobility/stupor, mutism, withdrawal) had a frequency of at least 50% and three factors (Increased, Abnormal, and Decreased Psychomotor Activity) accounted for 37% of the variance. One previous PCA of the BFCRS (Ungvari et al., 2007) identified four factors that explained 49.9% of the variance in patients with chronic schizophrenia. Their factors “Negative/Withdrawn” and “Agitated/Resistive” roughly correlate with our “Decreased Psychomotor Activity” and “Increased Psychomotor Activity” factors. Their third factor “Repetitive/Echo” is mostly subsumed under our “Increased Psychomotor Activity” factor. Finally, their “Automatic” factor roughly correlates with our “Abnormal Psychomotor Activity” factor. Autonomic Abnormality did not load onto any factor in either study, which may indicate a low specificity of this sign for catatonia. We show that the BFCRS includes many redundant items, but it is unclear at this point which items ought to be removed while retaining the structure of the BFCRS. Since the PCA only explained 37% of variance, however, catatonia likely presents in more complex ways than can be explained by three factors. The response option “frequent” may be combined with either “occasional” or “constant”, without loss of much information. This is unlike a prior analysis based on patients with chronic schizophrenia (Wong et al., 2007) that suggested collapsing the response options to present or absent only. Our study shows that the BFCRS has low reliability at low severity of catatonia, but good reliability at moderate–high severity of catatonia. This provides direction for future assessments of catatonia. 4.3. Limitations We used a clinical sample of acutely ill patients with a high degree of clinical suspicion for catatonia. This does not allow us to make any statements about the incidence or prevalence of catatonia. We were unable to assess inter-rater reliability, although all raters were psychiatrists or psychiatric trainees trained in the assessment of catatonia with the BFCRS. Our sample size was small, but larger than in previous PCA and IRT studies of catatonia. Furthermore, our sample was unique given the inclusion of both medical and non-psychotic patients and the application of an IRT technique not previously applied to the BFCRS. Our study has sources of heterogeneity that may not be generalizable. Prior study cohorts are largely Caucasian and one Chinese, while the current cohort is largely American Caucasians. Finally, the current study utilizes a retrospective rather than prospective study design. 5. Conclusions We show that the exclusion of staring may lower the detection of catatonia. Three components (Increased/Abnormal/Decreased Psychomotor Activity) still leave 63% of the variance in catatonia unexplained. The item responses of the BFCRS were infrequently affirmative (over 50% for only 4 items), superfluous, and reliable only at moderately high catatonia levels. We conclude that the structure of catatonia still needs to be discovered (45) and recommend that a consensus committee of catatonia scholars (16) designs an improved rating instrument for catatonia.
Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.schres.2014.12.036. Role of funding source The authors have no funding body agreements to disclose.
Contributors Dr. Wilson and Dr. Niu contributed equally to the manuscript by undertaking data management, literature review, creation of tables and figures and statistical analyses. Dr. Wilson and Dr. Niu were the main authors of the final text. Dr. Wilson and Dr. Niu should be recognized as joint first authors. Dr. Nicolson assisted with data management and interpretation, literature review and contributed to the final text. Dr. Levine oversaw the statistical analyses and contributed to the final text. Dr. Heckers, as the senior author, was responsible for the overall design of the study, interpretation of analyses and contributed significantly to the final version of the manuscript. All authors have contributed significantly to the paper and have approved of the final version of the manuscript. All authors agree to the order authorship and should be listed in the below order: Jo Ellen Wilson, MD; Kathy Niu, MD; Stephen E. Nicolson, MD; Stephen Z. Levine, PhD; and Stephan Heckers, MD MSc. Conflict of interest Dr. Wilson, Dr. Niu, Dr. Nicolson and Dr. Heckers report no competing interests. Dr. Levine discloses that he has received research support, and/or consultancy fees and/or travel support from F. Hoffmann-La Roche and Eli Lilly, however that work is unrelated to the current submission.
Acknowledgments The authors have no acknowledgments to make.
References APA, 2000. Diagnostic and Statistical Manual of Mental Disorders. IV ed. American Psychiatric Association. Abrams, R., Taylor, M.A., 1976. Catatonia. A prospective clinical study. Arch. Gen. Psychiatry 33 (5), 579–581. Abrams, R., Taylor, M.A., 1977. Catatonia: prediction of response to somatic treatments. Am. J. Psychiatry 134 (1), 78–80. Abrams, R., Taylor, M.A., Coleman Stolurow, K.A., 1979. Catatonia and mania: patterns of cerebral dysfunction. Biol. Psychiatry 14 (1), 111–117. APA, 2013. Diagnostic and Statistical Manual of Mental Disorders. 5 ed. Bleuler, E., 2010. Dementia Praecox or the Group of Schizophrenias. pp. 394–400. Braff, D.L., Ryan, J., Rissling, A.J., Carpenter, W.T., 2013. Lack of use in the literature from the last 20 years supports dropping traditional schizophrenia subtypes from DSM-5 and ICD-11. Schizophr. Bull. 39 (4), 751–753. Bush, G., Fink, M., Petrides, G., Dowling, F., Francis, A., 1996a. Catatonia. I. Rating scale and standardized examination. Acta Psychiatr. Scand. 93 (2), 129–136. Bush, G., Fink, M., Petrides, G., Dowling, F., Francis, A., 1996b. Catatonia. II. Treatment with lorazepam and electroconvulsive therapy. Acta Psychiatr. Scand. 93 (2), 137–143. Caroff, S.N., Mann, S.C., Campbell, E.C., Sullivan, K.A., 2002. Movement disorders associated with atypical antipsychotic drugs. J. Clin. Psychiatry 63 (Suppl. 4), 12–19. Fink, M., 2006. Catatonia: A Clinician's Guide to Diagnosis and Treatment. Cambridge University Press. Fink, M., Shorter, E., Taylor, M.A., 2010. Catatonia is not schizophrenia: Kraepelin's error and the need to recognize catatonia as an independent syndrome in medical nomenclature. Schizophr. Bull. 36 (2), 314–320. Francis, A., Fink, M., Appiani, F., Bertelsen, A., Bolwig, T.G., Braunig, P., Caroff, S.N., Carroll, B.T., Cavanna, A.E., Cohen, D., Cottencin, O., Cuesta, M.J., Daniels, J., Dhossche, D., Fricchione, G.L., Gazdag, G., Ghaziuddin, N., Healy, D., Klein, D., Kruger, S., Lee, J.W., Mann, S.C., Mazurek, M., McCall, W.V., McDaniel, W.W., Northoff, G., Peralta, V., Petrides, G., Rosebush, P., Rummans, T.A., Shorter, E., Suzuki, K., Thomas, P., Vaiva, G., Wachtel, L., 2010. Catatonia in diagnostic and statistical manual of mental disorders, fifth edition. J. ECT 26 (4), 246–247. Gorsuch, R.L., 1993. Factor Analysis 2ed. Lawrence Erlbaum Associates, New Jersey. Heckers, S., Tandon, R., Bustillo, J., 2010. Catatonia in the DSM—shall we move or not? Schizophr. Bull. 36 (2), 205–207. Kahlbaum, K., 1874. Die Katatonie oder das Spannungsirresein. Hirschwald, Berlin. Kraepelin, E., 1919. Dementia Praecox and Paraphrenia. Chicago Medical Book Co, Chicago. Kruger, S., Bagby, R.M., Hoffler, J., Braunig, P., 2003. Factor analysis of the catatonia rating scale and catatonic symptom distribution across four diagnostic groups. Compr. Psychiatry 44 (6), 472–482. Kruskal, J.B., Wish, M., 1978. Multidimensional Scaling. Sage, Newbury, CA. Levine, S.Z., Rabinowitz, J., Rizopoulos, D., 2011. Recommendations to improve the positive and negative syndrome scale (PANSS) based on item response theory. Psychiatry Res. 188 (3), 446–452.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx Lopez-Canino, A., F.A., 2004. Drug-induced catatonia. In: Caroff, S.N., M.S., Francis, A., Fricchione, G.L. (Eds.), Catatonia: From Psychopathology to Neurobiology. American Psychiatric Press, Inc., Washington, D.C., pp. 129–139. McKenna, P.J., Lund, C.E., Mortimer, A.M., Biggins, C.A., 1991. Motor, volitional and behavioural disorders in schizophrenia. 2: the ‘conflict of paradigms’ hypothesis. Br. J. Psychiatry 158, 328–336. Morrison, J.R., 1973. Catatonia. Retarded and excited types. Arch. Gen. Psychiatry 28 (1), 39–41. Morrison, J.R., 1974a. Catatonia: prediction of outcome. Compr. Psychiatry 15 (4), 317–324. Morrison, J.R., 1974b. Karl Kahlbaum and catatonia. Compr. Psychiatry 15 (4), 315–316. Morrison, J.R., 1975. Catatonia: diagnosis and management. Hosp. Community Psychiatry 26 (2), 91–94. Northoff, G., Koch, A., Wenke, J., Eckert, J., Boker, H., Pflug, B., Bogerts, B., 1999. Catatonia as a psychomotor syndrome: a rating scale and extrapyramidal motor symptoms. Mov. Disord. 14 (3), 404–416. Oulis, P., Lykouras, L., Tomaras, V., Panayotopoulou, V., Gournellis, R., Stefanis, C., 1997. DSM-IV catatonic features among psychiatric inpatients: a preliminary study. Eur. Psychiatry 12 (8), 412–414. Peralta, V., Cuesta, M.J., 2001. Motor features in psychotic disorders. I. Factor structure and clinical correlates. Schizophr. Res. 47 (2–3), 107–116. Peralta, V., Cuesta, M.J., Mata, I., Serrano, J.F., Perez-Nievas, F., Natividad, M.C., 1999. Serum iron in catatonic and noncatatonic psychotic patients. Biol. Psychiatry 45 (6), 788–790. Peralta, V., Campos, M.S., de Jalon, E.G., Cuesta, M.J., 2010. DSM-IV catatonia signs and criteria in first-episode, drug-naive, psychotic patients: psychometric validity and response to antipsychotic medication. Schizophr. Res. 118 (1–3), 168–175.
7
Rizopoulos, D., 2006. ltm: an R package for latent variable modelling and item response theory analyses. J. Stat. Softw. 17 (5), 1–25. Rosebush, P.I., Mazurek, M.F., 2010. Catatonia and its treatment. Schizophr. Bull. 36 (2), 239–242. Sienaert, P., Rooseleer, J., De Fruyt, J., 2011. Measuring catatonia: a systematic review of rating scales. J. Affect. Disord. 135 (1–3), 1–9. Starkstein, S.E., Petracca, G., Teson, A., Chemerinski, E., Merello, M., Migliorelli, R., Leiguarda, R., 1996. Catatonia in depression: prevalence, clinical correlates, and validation of a scale. J. Neurol. Neurosurg. Psychiatry 60 (3), 326–332. Tabachnick, B.G., Fidell, L.S., 2001. Using Multivariate Statistics. 4 ed. Harper & Colins, NY. Tandon, R., Heckers, S., Bustillo, J., Barch, D.M., Gaebel, W., Gur, R.E., Malaspina, D., Owen, M.J., Schultz, S., Tsuang, M., van Os, J., Carpenter, W., 2013. Catatonia in DSM-5. Schizophr. Res. 150, 26–30. Taylor, M.A., Abrams, R., 1977. Catatonia. Prevalence and importance in the manic phase of manic-depressive illness. Arch. Gen. Psychiatry 34 (10), 1223–1225. Taylor, M.A., Fink, M., 2003. Catatonia in psychiatric classification: a home of its own. Am. J. Psychiatry 160 (7), 1233–1241. Ungvari, G.S., Goggins, W., Leung, S.K., Gerevich, J., 2007. Schizophrenia with prominent catatonic features (‘catatonic schizophrenia’). II. Factor analysis of the catatonic syndrome. Prog. Neuropsychopharmacol Biol. Psychiatry 31 (2), 462–468. Vesperini, S., Papetti, F., Pringuey, D., 2010. Are catatonia and neuroleptic malignant syndrome related conditions? Encéphale 36 (2), 105–110. Wong, E., Ungvari, G.S., Leung, S.K., Tang, W.K., 2007. Rating catatonia in patients with chronic schizophrenia: Rasch analysis of the Bush-Francis Catatonia Rating Scale. Int. J. Methods Psychiatr. Res. 16 (3), 161–170.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
Contents lists available at ScienceDirect
Schizophrenia Research journal homepage: www.elsevier.com/locate/schres
The diagnostic criteria and structure of catatonia Jo Ellen Wilson a,⁎,1, Kathy Niu a,1, Stephen E. Nicolson a, Stephen Z. Levine b, Stephan Heckers a a b
Department of Psychiatry, Vanderbilt University, Nashville, TN, USA Department of Community Mental Health, University of Haifa, Israel
a r t i c l e
i n f o
Article history: Received 31 January 2014 Received in revised form 18 December 2014 Accepted 22 December 2014 Available online xxxx Keywords: Catatonia DSM-IV DSM-5 Diagnosis Principal component analysis Item response theory
a b s t r a c t Background: The classification of catatonia has fluctuated and underwent recent changes in DSM-5. The current study examines the prevalence of catatonia signs, estimates the utility of diagnostic features, identifies core catatonia signs, and explores their underlying structure. Method: We screened 339 acutely ill medical and psychiatric patients with the Bush Francis Catatonia Rating Scale (BFCRS). We examined prevalence and severity of catatonia signs and compared BFCRS, DSM-IV and DSM-5 diagnoses. We used principal component analysis (PCA) to examine the factorial validity of catatonia and item response theory (IRT) to estimate each sign's utility and reliability. Results: Out of the 339 patients, 300 were diagnosed with catatonia using the BFCRS and 232 catatonia diagnoses were validated by the treating provider based on selection for treatment with benzodiazepines or electroconvulsive therapy. Of the 232 validated catatonia cases, 211 (91%) met DSM-IV criteria but only 170 (73%) met DSM-5 criteria for catatonia. Staring was the most prevalent catatonia sign. PCA identified three components, interpretable as “Increased, Abnormal and Decreased Psychomotor Activity,” although 63% of the variance was unexplained. IRT showed that Excitement, Waxy Flexibility and Immobility/Stupor were the best indicators of each factor. The BFCRS had many redundant items and as a whole had low reliability at low severity of catatonia, but good reliability at moderate–high severity of catatonia. Conclusions: The structure of catatonia remains to be discovered. Published by Elsevier B.V.
1. Introduction In 1874, Karl Kahlbaum described catatonia in patients who suffered from severe psychotic, mood and medical conditions (Kahlbaum, 1874). Kraepelin and Bleuler, however, redefined catatonia as a subtype of dementia praecox (1896) (Kraepelin, 1919) and schizophrenia (1911) (Bleuler, 2010). Clinicians followed their redefinition throughout most of the 20th century, with a few notable exceptions (Morrison, 1973, 1974a,b, 1975; Abrams and Taylor, 1976, 1977; Taylor and Abrams, 1977; Abrams et al., 1979). The diagnostic categories for catatonia were broadened in DSM-III-R (catatonia was added as a specifier for mood disorders) and DSM-IV (catatonia resulting from a general medical condition was added) (AP, 2000). DSM-5 introduced several changes in the classification of catatonia, intended to improve recognition, treatment and research (Taylor and Fink, 2003; Fink et al.,
⁎ Corresponding author at: Vanderbilt Psychiatric Hospital, 1601 23rd Ave. South, Nashville, TN 37212, USA. Tel.: +1 615 936 3555. E-mail addresses: [email protected] (J.E. Wilson), [email protected] (K. Niu), [email protected] (S.E. Nicolson), [email protected] (S.Z. Levine), [email protected] (S. Heckers). 1 Dr. Wilson and Dr. Niu are joint first authors.
2010; Francis et al., 2010; Heckers et al., 2010; Rosebush and Mazurek, 2010; Braff et al., 2013; Tandon et al., 2013). Despite the changes in nosology, it is unclear which and how many signs define a case of catatonia. More than 40 clinical signs of catatonia have been described and at least seven scales for the assessment of catatonia have been proposed (Taylor and Fink, 2003). The preferred rating scale for routine catatonia assessment is the 23-item BushFrancis Catatonia Rating Scale (BFCRS) because of reports of its validity and reliability, ease of administration and since cases can be identified based on the presence of 2 or more items of the first 14 items (Sienaert et al., 2011). In contrast, the DSM-5 criteria require a minimum of 3 out of 12 signs for the diagnosis of catatonia. How these differences in the definition of caseness affect the recognition of catatonia is currently unknown. Previous studies have identified up to six dimensions of catatonia (Abrams et al., 1979; McKenna et al., 1991; Starkstein et al., 1996; Oulis et al., 1997; Northoff et al., 1999; Peralta et al., 1999, 2010; Peralta and Cuesta, 2001; Kruger et al., 2003; Ungvari et al., 2007). Except for one study (Oulis et al., 1997), the factors stupor and excitement are a common finding, although the naming and exact grouping of signs differ. Here we report the analysis of a diverse clinical sample of acutely ill medical and psychiatric patients with catatonia. We compare diagnosis
http://dx.doi.org/10.1016/j.schres.2014.12.036 0920-9964/Published by Elsevier B.V.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
2
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
rates based on BFCRS, DSM-IV and DSM-5, study the factorial structure of catatonia and estimate the reliability of the catatonia signs in the BFCRS. 2. Methods 2.1. Registry Patients were identified using a retrospective chart review. We included all patients with a completed Bush Francis Catatonia Rating Scale (BFCRS) assessment in the electronic medical record system at Vanderbilt University Medical Center between 06/18/2009 and 06/16/ 2012. All catatonia assessments were performed by psychiatrists who had experience in assessing catatonia and were trained in the use of the BFCRS. We identified 339 patients with a BFCRS assessment performed in one of four settings: general medical hospital, emergency room, psychiatric hospital or outpatient psychiatry clinic. If a patient was represented with more than one rating scale, only the first assessment performed by the most senior evaluator was included for analysis. All 339 patients were screened for catatonia with the 14-item Bush-Francis Catatonia Screening Instrument (BFCSI). 300 patients (88.5%) scored for 2 or more items and were assessed with the full 23-item BFCRS (see Supplemental Fig. 1). Careful case-by-case review of all available medical records was undertaken to determine the underlying medical and psychiatric diagnoses (see Supplemental Fig. 2). Additionally, response to treatment with benzodiazepines and ECT was scored as “substantially improved”, “partially improved” or “not improved”, as documented in the notes by the treating physician. 2.2. BFCSI and BFCRS The Bush Francis Catatonia Rating Scale (BFCRS) is the most widely used catatonia rating scale, given its ease of use and high inter-rater reliability and validity (Sienaert et al., 2011). The scale consists of 23 signs, with the first 14 items constituting the Bush Francis Catatonia Screening Instrument (BFCSI). The definition of catatonia proposed by Bush et al. (1996a) is the presence of two or more signs on the BFCSI, regardless of severity (Bush et al., 1996a,b). Sixteen items are rated on a severity scale 0 (absent) through 3 (e.g. constant), while the other six items are rated 0 (absent) and 3 (present). The total BFCRS score is the sum of the responses to all 23 items. 2.3. Catatonia diagnoses The presence of 2 or more BFCSI signs on examination meets the standard for a diagnosis of catatonia (Bush et al., 1996a,b). Within the 300 patients who met the BFCSI diagnosis of catatonia, we indentified a subset of 232 patients meeting the following 3 criteria: 1) at least 2 items on BFCSI, 2) clinical diagnosis of catatonia by treating provider and 3) selection for treatment with a benzodiazepine or ECT. We will focus our report on this subset of 232 catatonia patients validated by treatment. To determine caseness for catatonia we calculated the number of positive items on the BFCSI for each subject and identified which signs, if removed, lowered the sum below 2. DSM-IV and DSM-5 diagnoses of catatonia were established by applying the respective diagnostic criteria A (APA, 2013) to all cases with 2 or more BFCSI signs. To ascertain sensitivity of diagnostic criteria we examined the severity and prevalence of catatonic signs, catatonia caseness, the effect of treatment and the potential confounding effect of psychopharmacological treatment history. 2.4. Psychometric analysis Data analysis proceeded in three steps. Step one consisted of computing item (i.e., sign) frequencies. In step two, principal components analysis (PCA) (Kruskal and Wish, 1978; Gorsuch, 1993;
Tabachnick and Fidell, 2001) was computed to identify the number of dimensions that underlie the BFCRS. To identify the most parsimonious number of dimensions, a scree plot was generated, and results from 1000 simulations of equal data size and symptoms were superimposed on the graph. This was supplemented by considerations for reliability and interpretability. Promax rotation allows for correlation between dimensions and was used to rotate the structural coefficients to interpret the dimensions (Gorsuch, 1993). The cut-off value for item loadings was 0.40. In the third step, item response theory (IRT) analysis was computed with Latent Trait Models (LTM) (Rizopoulos, 2006) (Levine et al., 2011). IRT quantifies the disparity between clinical ratings and the expression of the underlying (latent) catatonia phenotype.
3. Results 3.1. Demographic and clinical characteristics Patients were assessed with the BFCSI in a psychiatric hospital (n = 218, 64.3%), a tertiary medical center (n = 66, 19.5%), an emergency room (n = 44, 13.0%) and a psychiatry outpatient clinic (n = 11, 3.2%) (Supplemental Table 1).
3.2. Severity and prevalence of catatonia signs In the group of 339 patients screened, we observed a mean of 5.27 (2.39 SD) signs using the BFCSI and 7.58 (3.51 SD) signs using the full BFCRS. Staring was the most prevalent sign (observed in 70.5% of all patients) and combativeness was the least frequently reported sign (observed in 10.9% of all patients) (Fig. 1A). 20 patients (8.6% of all patients) showed only 2 signs, i.e., the minimum number of signs for caseness (see Fig. 1B). For the 232 validated cases of catatonia, the mean BFCSI and BFCRS scores were 5.60 ± 2.34 and 8.08 ± 3.45 respectively.
3.3. Catatonia caseness We assessed how often cases would fail to meet the criteria for catatonia if an item was removed from the criterion set for the BFCSI, DSM-IV and DSM-5 (Fig. 1C). Staring had the greatest impact on caseness, followed by immobility/stupor, excitement, rigidity and mutism. Grimacing, echopraxia/echolalia and waxy flexibility had no effect on caseness. Of the 232 patients with validated catatonia, 211 (90.95%) met DSM-IV criteria but only 170 (73.28%) met DSM-5 criteria for catatonia.
3.4. Effect of treatment with benzodiazepines and ECT Of the 232 cases selected for treatment with benzodiazepines or ECT, 7 (3%) patients refused the recommended treatment. The average effective dose of lorazepam required for treatment response was 5.81 mg per day (SD 4.7). 6 patients (2.6%) were treated with clonazepam, requiring an average dose of 2.8 mg per day until clinical improvement. Of the 225 patients treated with a benzodiazepine, 191 (84.9%) substantially improved, 25 (11.1%) partially improved, 9 (3.1%) did not respond to treatment with a benzodiazepine. 45 (19.4%) patients required treatment with ECT, generally due to the presence of severe catatonia (autonomic abnormalities, refusal to eat or drink), or incomplete response to benzodiazepines. 42 patients (93.3% of those who received ECT) substantially improved, while 3 patients (6.7%) partially improved. The average number of ECT sessions required during induction for treatment remission was 8.4 (SD 5.6). 1 patient was already undergoing maintenance ECT at the time catatonia was assessed.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
Staring Immobility/Stupor Musm Withdrawal Posturing/Catalepsy Negavism Automac Obedience Rigidity Autonomic Abnormality Ambitendency Mitgehen Stereotypy Impulsivity Perseveraon Excitement Grimacing Verbigeraon Waxy Flexibility Gegenhalten Echopraxia/Echolalia Mannerisms Grasp reflex Combaveness
80% 65% 56% 49% 45% 44% 41% 38% 32% 31% 28% 27% 27% 26% 24% 24% 21% 20% 20% 18% 15% 14% 12%
B
C 18%
Waxy Flexibility Echopraxia/echolalia Grimacing Verbigeraon Withdrawal Negavism Stereotypy Mannerisms Posturing/catalepsy Rigidity Musm Excitement Imobility/Stupor Staring
16% 14%
BFCSI BFCRS
Percentage of Paents
A
3
12% 10% 8% 6% 4% 2% 0% 0
1
2
3
Frequency of catatonia signs
4
5
6
7
8
9 10 11 12 13 14
Number of catatonia signs
0
1
2
3
4
5
Caseness of catatonia signs (%)
Fig. 1. Prevalence, severity and caseness of catatonia signs. A. Prevalence of catatonia signs in 300 patients with a score of ≥2 on the BFCSI. BFCSI items are shown in black, remaining BFCRS items are shown in gray. B. Severity of catatonia, as measured by the number of catatonia signs in patients screened with the BFCSI. C. Effect of caseness after removal of individual BFCSI items.
3.5. Effect of treatment with antipsychotic medication 103 (44.4%) of the 232 patients with validated catatonia were being treated with either a typical or atypical antipsychotic medication at the time of the assessment (Table 1). Treated and untreated patients did not differ in the mean number of catatonia signs, but the prevalence of various items differed (see Supplemental Fig. 3). Patients treated with an antipsychotic drug showed significantly (p b 0.05) less mutism, staring, grimacing, withdrawal and significantly (p b 0.05) more automatic obedience, mitgehen and impulsivity. 3.6. Prevalence of catatonia signs Table 2 shows the prevalence of the catatonia signs in the 232 patients meeting BFCRS criteria. Only four symptoms exceeded a frequency of 50%: Staring (N = 189, 81.5%), Immobility/Stupor (N = 162, 69.8%), Mutism (N = 144, 62.1%) and Withdrawal (N = 122, 52.6%). For all items, the most severe rating of “3” was the least selected rating option. For Stereotypy, none responded with a “3.” The four symptoms rated most frequently as absent were Combativeness (N = 206, 88.79%), Grasp Reflex (N = 200, 86.21%), Mannerisms (N = 194, 83.62%) and Stereotypy (N = 186, 80.17%). 3.7. Psychometric analysis: principle component analysis (PCA) Here we report the psychometric analysis (PCA and IRT) of the 232 validated catatonia cases.
A three-factor solution explained 37% of the total variance (Table 3 and Supplemental Fig. 4). Factor 1 consisted of Excitement, Impulsivity, Combativeness, Verbigeration, Mannerisms, Grimacing, and Stereotypy, and was hence labeled “Increased Psychomotor Activity.” Factor 2 consisted of Waxy Flexibility, Automatic Obedience, Mitgehen, Ambitendency, Grasp Reflex, Gegenhalten, and Posturing/Catalepsy, and was hence labeled “Abnormal Psychomotor Activity”. Factor 3 consisted of Immobility/Stupor, Mutism, Withdrawal, Staring, and Negativism, and was hence labeled “Decreased Psychomotor Activity”. Perseveration, Echopraxia/Echolalia, Autonomic Abnormality, and Rigidity did not load on a factor. The three factors showed moderate correlations at best, the highest between factors 3 (Decreased Psychomotor Activity) and 2 (Abnormal Psychomotor Activity) (correlation 0.241). PCA of the 300 cases meeting BFCRS criteria for catatonia also revealed 3 factors, but the cumulative variance ranking was slightly different with Increased Psychomotor Activity (15.09%), Decreased Psychomotor Activity (28.75%) and Abnormal Psychomotor Activity (35.96%). 3.8. Psychometric analysis: item response theory (IRT) Item response theory (IRT) was computed for each of the BFCRS factors to test the strength of each item's relationship between ratings and the latent phenotype catatonia (Supplemental Table 1). To quantify this relationship, we defined a discrimination parameter value (α) of 1 as a data-driven cut-off value. The Factor 1 items with
Table 1 Demographic and clinical characteristics.
Demographics Age Gender (F/M) Race (Caucasian/AA/other) Education Clinical characteristics Psychiatric hospital Medical center Emergency room Psychiatry Clinic Number of BFCSI items present Number of BFCRS items present Chlorpromazine equivalent; % of subjects treated
Patients with catatonia, validated by treatment n = 232
Patients with catatonia, not validated by treatment n = 68
Patients without catatonia n = 39
38.37 ± 18.68 128/104 151/65/16 12.16 ± 2.98
36.10 ± 18.12 34/34 39/22/7 11.31 ±2.79
35.54 ± 18.89 16/23 20/16/3 11.92 ± 2.53
153 (65.95) 44 (18.97) 27 (11.64) 8 (3.45) 5.6 ± 2.34 8.08 ± 3.45 173.26 ± 376.08; (44.40)
43 (63.24) 13 (19.12) 11 (16.18) 1 (1.47) 4.12 ±2.24 5.97 ±3.26 90.66 ±191.06; (33.82)
22 (56.41) 9 (23.08) 6 (15.38) 2 (5.13) 0.74 ± 0.64 2.28 ± 2.80 184.34 ± 277.4; (51.28)
Numbers are mean ± SD. Percentage of sample in parentheses.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
4
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
Table 2 BFCRS item responses in 232 catatonia patients. Response
1. Excitement 2. Immobility/Stupor 3. Mutism 4. Staring 5. Posturing/Catalepsy 6. Echopraxia/Echolalia 7. Stereotypy 8. Grimacing 9. Mannerisms 10. Verbigeration 11. Rigidity 12. Negativism 13. Waxy Flexibility 14. Withdrawal 15. Impulsivity 16. Automatic Obedience 17. Mitgehen 18. Gegenhalten 19. Ambitendency 20. Grasp Reflex 21. Perseveration 22. Combativeness 23. Autonomic Abnormality
0 e.g. “absent”
1 e.g. “occasional”
2 e.g. “frequent”
3 e.g. “constant”
N
(%)
N
N
N
179 71 90 41 120 175 186 162 194 179 142 127 183 110 168 137
(77.16) 37 (15.95) 14 (6.03) 2 (.86) (30.60) 127 (54.74) 33 (14.22) 1 (.43) (38.79) 52 (22.41) 66 (28.45) 24 (10.34) (17.67) 91 (39.22) 88 (37.93) 12 (5.17) (51.72) 54 (23.28) 46 (19.83) 12 (5.17) (75.43) 41 (17.67) 12 (5.17) 4 (1.72) (80.17) 28 (12.07) 18 (7.76) 0 (.00) (69.83) 31 (13.36) 32 (13.79) 7 (3.02) (83.62) 26 (11.21) 9 (3.88) 3 (1.29) (77.16) 24 (10.34) 24 (10.34) 5 (2.16) (61.21) 55 (23.71) 30 (12.93) 5 (2.16) (54.74) 60 (25.86) 32 (13.79) 13 (5.60) (78.88) 0 (.00) 0 (.00) 49 (21.12) (47.41) 41 (17.67) 69 (29.74) 12 (5.17) (72.41) 39 (16.81) 23 (9.91) 2 (.86) (59.05) 30 (12.93) 36 (15.52) 29 (12.50)
158 180 150 200 165 206 156
(68.10) (77.59) (64.66) (86.21) (71.12) (88.79) (67.24)
(%)
(%)
0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 0 (.00) 17 (7.33) 7 (3.02) 45 (19.40) 25 (10.78)
74 52 82 32 67 2 6
(%)
(31.90) (22.41) (35.34) (13.79) (28.88) (.86) (2.59)
an α exceeding 2 were Excitement, Impulsivity, Combativeness, Mannerisms, indicating that these were the most representative items of the “Increased Psychomotor Activity” dimension. All other loaded items — Verbigeration, Grimacing, and Stereotypy had good discrimination with α N 1 as well. Factor 2 had Waxy Flexibility with α N 2; Gegenhalten, Grasp Reflex, Mitgehen with α N 1, and Ambitendency and Posturing/Cataplexy close by. Automatic Table 3 PCA factors, loading values, and factor correlation matrix. Component
Excitement Impulsivity Combativeness Verbigeration Mannerisms Grimacing Stereotypy Perseveration Echopraxia/Echolalia Waxy Flexibility Automatic Obedience Mitgehen Ambitendency Grasp Reflex Gegenhalten Posturing/Catalepsy Autonomic Abnormality Immobility/Stupor Mutism Withdrawal Staring Negativism Rigidity Cumulative % of variance Eigenvalues Reliability Kaiser–Meyer–Olkin measure
1
2
3
0.8 0.77 0.74 0.61 0.6 0.5 0.41 0.35 0.26 −0.03 −0.08 −0.03 −0.16 0.24 0.09 −0.09 0.04 −0.33 −0.02 0.1 −0.17 0.17 0.26 16.11% 3.71 0.77 0.73
−0.13 −0.12 −0.19 0.08 −0.05 0.19 0.32 0.2 0.19 0.65 0.55 0.54 0.53 0.47 0.46 0.46 0.16 −0.04 −0.06 −0.22 0.26 0.15 0.35 29.53% 3.09 0.61
−0.13 0.1 0.24 0.05 −0.15 0 −0.16 −0.19 0.21 0.02 −0.19 0.02 −0.16 0.02 0.11 0.36 0.08 0.71 0.71 0.71 0.45 0.43 0.35 36.99% 1.72 0.64
Numbers in bold represent BFCRS items that loaded onto the respective catatonia component.
Obedience had the least strong relationship to the “Abnormal Psychomotor Activity” dimension. Factor 3 had Immobility/Stupor with α N 2; Mutism, Withdrawal, and Staring with α N1. However, Negativism had the least strong relationship to the “Decreased Psychomotor Activity” dimension. Threshold parameters (βs) indicate the severity of catatonia at which two rating response categories are equally likely to be selected. For example, β1 represents equal likelihood of selecting response 1 “absent” and response 2 “occasional.” Of note, only items in Factor 3 had a β b − 1, allowing for more information at low levels of catatonia. If a threshold value exceeds 1.96, it suggests ratings provide good information. The Item Response Category Characteristic Curves (ICC) graphically represents a combination of the item discrimination and difficulty, where the probability of a certain response is graphed against the severity of catatonia (left column of Fig. 2 and Supplemental Fig. 5). Examination of the ICCs shows that Verbigeration, Grimacing, Automatic Obedience, Withdrawal have substantial overlap between rating options “1” and “2,” suggesting that these two rating options could be collapsed into one without substantial data loss. Notably, Negativism shows poor separation of response choices, suggesting that the choices between mild, moderate, and severe resistance or contrary behavior is not informative to distinguish the severity of Factor 3. The information curves show how much information or reliability an item offers relative to the severity of catatonia (right column of Fig. 2). Most items peak around the same range of catatonia severity, which suggests redundancy. The individual item information curves combine to form the test information curve for that set of items. The test information curve shows that the set of items for Factor 1 is reliable for mostly mid-high level of “Increased Psychomotor Activity” catatonia, Factor 2 items are reliable at mid-high levels of “Abnormal Psychomotor Activity” catatonia, and Factor 3 items are reliable at low, medium, and high levels of “Decreased Psychomotor Activity” catatonia. No set of items are reliable at very low levels of catatonia. 4. Discussion We assessed the prevalence of catatonia signs in a large clinical sample, tested the effect of different criteria sets on categorical diagnoses and explored its core factor structure. Our results suggest that the DSM-5 criteria are more restrictive than previous criteria sets, excluding more than a quarter of patients meeting BFCRS criteria from the diagnosis. Principal component analysis identified three factors labeled “Increased Psychomotor Activity”, “Abnormal Psychomotor Activity”, and “Decreased Psychomotor Activity” that cumulatively accounted for 37% of total variance. Item response theory identified that the BFCRS lacks reliability at low levels of catatonia. 4.1. Effect of diagnostic criteria on recognition of catatonia Staring is recognized as one of the most prevalent catatonia signs, with estimates ranging from 25% to 92% (18, 47–57). This was confirmed in our sample, where staring was the most prevalent catatonia sign. Yet, staring is not included in any version of the DSM. We recommend that staring be included in the diagnostic criteria for catatonia, since it can be a crucial diagnostic sign (in 4% of our patients meeting BFCSI criteria). Rigidity is recognized by most catatonia rating scales (Sienaert et al., 2011) and in earlier versions of the DSM, but was removed from the DSM-5 criteria. This is supported by our analysis, since removal of rigidity affected caseness in less than 2% of cases. 44.4% of our patients were treated with antipsychotic medication at the time of the assessment. While the severity of catatonia did not differ between treated and untreated patients, some signs were more or less prevalent in the treated patients. Antipsychotic medication may induce an extrapyramidal syndrome akin to catatonia, which may exist on a
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
Item Response Category Characterisc Curves
Item Informaon Curves
1.5
excitement impulsivity combativeness mannerisms verbigeration grimacing stereotypy
0.0
0.0
0.5
1.0
0 1 2 3
Informaon
0.6
Rating
0.4
2.0
0.8
2.5
1.0
Informaon Test Curves: Factor 1
0.2
Probability
Increased Psychomotor Acvity
5
−4
−2
0
2
−4
4
−2
Excitement
0
Catatonia
2
4
1.0
waxyflexibility automaticobedience mitgehen ambitendency graspreflex gegenhalten posturingcatalepsy
0.5
Informaon
0 1 2 3
0.4
0.6
Rating
0.0
0.0
0.2
Probability
Decreased Psychomotor Acvity
0.8
1.5
1.0
Informaon Test Curves: Factor 2
−4 −4
−2
0
Musm
2
−2
4
0
2
4
Catatonia
2.0
mutism immobilitystupor withdrawal negativism staring
0.5
1.0
1.5
Informaon
0.8
0 3
0.4
0.6
Rating
0.0
0.0
0.2
Probability
Abnormal Psychomotor Acvity
2.5
1.0
3.0
Informaon Test Curves: Factor 3
−4
−2
0
2
Waxy Flexibility
4
−4
−2
0
2
4
Catatonia
Fig. 2. Item response category characteristic curves and information curves. Item response category characteristic curves (left column) are a probabilistic index of the relationship between ratings for an item (i.e., catatonia sign) and level on the latent construct (i.e., catatonia). For example, the curve for the sign “Excitement” depicts the distinction between “absent” that is likely at low symptom severity and “two” that is likely at higher symptom severity. Item information curves (right column) index items over latent severity to assess reliability at different severity levels. For example, catatonia Factor 1 (Increased Psychomotor Activity) assesses little catatonia when the item is low (x-axis values −4 to 0), more at moderate levels (x axis values 0 to 2), and little at high levels (x axis values 2 to 4).
spectrum between simple non-malignant catatonia and malignant catatonia or NMS (Caroff et al., 2002; Vesperini et al., 2010). Although the exact incidence of drug-induced catatonia is unknown, certain catatonia signs are associated with antipsychotic medication, including akinesia/stupor, mutism, catalepsy and waxy flexibility (Lopez-Canino
and F.A., 2004). Due to the retrospective design, the authors can only speculate about the effect of neuroleptic treatment. We assume that providers were more apt to reduce antipsychotic drug exposure in those with classic catatonic features (i.e., the Kahlbaum subtype of catatonia) (Fink, 2006), thus accounting for a significantly lower
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
6
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx
prevalence of mutism, staring, grimacing and withdrawal in the neuroleptic treated group. Additionally, we speculate that medical providers might have been inclined to provide a neuroleptic to an agitated patient either due to concern for violence or because of lack of recognition of the catatonic syndrome, therefore accounting for a higher prevalence of automatic obedience, mitgehen and impulsivity in the neuroleptic treated group. 4.2. Psychometric properties of BFCRS Only four items (staring, immobility/stupor, mutism, withdrawal) had a frequency of at least 50% and three factors (Increased, Abnormal, and Decreased Psychomotor Activity) accounted for 37% of the variance. One previous PCA of the BFCRS (Ungvari et al., 2007) identified four factors that explained 49.9% of the variance in patients with chronic schizophrenia. Their factors “Negative/Withdrawn” and “Agitated/Resistive” roughly correlate with our “Decreased Psychomotor Activity” and “Increased Psychomotor Activity” factors. Their third factor “Repetitive/Echo” is mostly subsumed under our “Increased Psychomotor Activity” factor. Finally, their “Automatic” factor roughly correlates with our “Abnormal Psychomotor Activity” factor. Autonomic Abnormality did not load onto any factor in either study, which may indicate a low specificity of this sign for catatonia. We show that the BFCRS includes many redundant items, but it is unclear at this point which items ought to be removed while retaining the structure of the BFCRS. Since the PCA only explained 37% of variance, however, catatonia likely presents in more complex ways than can be explained by three factors. The response option “frequent” may be combined with either “occasional” or “constant”, without loss of much information. This is unlike a prior analysis based on patients with chronic schizophrenia (Wong et al., 2007) that suggested collapsing the response options to present or absent only. Our study shows that the BFCRS has low reliability at low severity of catatonia, but good reliability at moderate–high severity of catatonia. This provides direction for future assessments of catatonia. 4.3. Limitations We used a clinical sample of acutely ill patients with a high degree of clinical suspicion for catatonia. This does not allow us to make any statements about the incidence or prevalence of catatonia. We were unable to assess inter-rater reliability, although all raters were psychiatrists or psychiatric trainees trained in the assessment of catatonia with the BFCRS. Our sample size was small, but larger than in previous PCA and IRT studies of catatonia. Furthermore, our sample was unique given the inclusion of both medical and non-psychotic patients and the application of an IRT technique not previously applied to the BFCRS. Our study has sources of heterogeneity that may not be generalizable. Prior study cohorts are largely Caucasian and one Chinese, while the current cohort is largely American Caucasians. Finally, the current study utilizes a retrospective rather than prospective study design. 5. Conclusions We show that the exclusion of staring may lower the detection of catatonia. Three components (Increased/Abnormal/Decreased Psychomotor Activity) still leave 63% of the variance in catatonia unexplained. The item responses of the BFCRS were infrequently affirmative (over 50% for only 4 items), superfluous, and reliable only at moderately high catatonia levels. We conclude that the structure of catatonia still needs to be discovered (45) and recommend that a consensus committee of catatonia scholars (16) designs an improved rating instrument for catatonia.
Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.schres.2014.12.036. Role of funding source The authors have no funding body agreements to disclose.
Contributors Dr. Wilson and Dr. Niu contributed equally to the manuscript by undertaking data management, literature review, creation of tables and figures and statistical analyses. Dr. Wilson and Dr. Niu were the main authors of the final text. Dr. Wilson and Dr. Niu should be recognized as joint first authors. Dr. Nicolson assisted with data management and interpretation, literature review and contributed to the final text. Dr. Levine oversaw the statistical analyses and contributed to the final text. Dr. Heckers, as the senior author, was responsible for the overall design of the study, interpretation of analyses and contributed significantly to the final version of the manuscript. All authors have contributed significantly to the paper and have approved of the final version of the manuscript. All authors agree to the order authorship and should be listed in the below order: Jo Ellen Wilson, MD; Kathy Niu, MD; Stephen E. Nicolson, MD; Stephen Z. Levine, PhD; and Stephan Heckers, MD MSc. Conflict of interest Dr. Wilson, Dr. Niu, Dr. Nicolson and Dr. Heckers report no competing interests. Dr. Levine discloses that he has received research support, and/or consultancy fees and/or travel support from F. Hoffmann-La Roche and Eli Lilly, however that work is unrelated to the current submission.
Acknowledgments The authors have no acknowledgments to make.
References APA, 2000. Diagnostic and Statistical Manual of Mental Disorders. IV ed. American Psychiatric Association. Abrams, R., Taylor, M.A., 1976. Catatonia. A prospective clinical study. Arch. Gen. Psychiatry 33 (5), 579–581. Abrams, R., Taylor, M.A., 1977. Catatonia: prediction of response to somatic treatments. Am. J. Psychiatry 134 (1), 78–80. Abrams, R., Taylor, M.A., Coleman Stolurow, K.A., 1979. Catatonia and mania: patterns of cerebral dysfunction. Biol. Psychiatry 14 (1), 111–117. APA, 2013. Diagnostic and Statistical Manual of Mental Disorders. 5 ed. Bleuler, E., 2010. Dementia Praecox or the Group of Schizophrenias. pp. 394–400. Braff, D.L., Ryan, J., Rissling, A.J., Carpenter, W.T., 2013. Lack of use in the literature from the last 20 years supports dropping traditional schizophrenia subtypes from DSM-5 and ICD-11. Schizophr. Bull. 39 (4), 751–753. Bush, G., Fink, M., Petrides, G., Dowling, F., Francis, A., 1996a. Catatonia. I. Rating scale and standardized examination. Acta Psychiatr. Scand. 93 (2), 129–136. Bush, G., Fink, M., Petrides, G., Dowling, F., Francis, A., 1996b. Catatonia. II. Treatment with lorazepam and electroconvulsive therapy. Acta Psychiatr. Scand. 93 (2), 137–143. Caroff, S.N., Mann, S.C., Campbell, E.C., Sullivan, K.A., 2002. Movement disorders associated with atypical antipsychotic drugs. J. Clin. Psychiatry 63 (Suppl. 4), 12–19. Fink, M., 2006. Catatonia: A Clinician's Guide to Diagnosis and Treatment. Cambridge University Press. Fink, M., Shorter, E., Taylor, M.A., 2010. Catatonia is not schizophrenia: Kraepelin's error and the need to recognize catatonia as an independent syndrome in medical nomenclature. Schizophr. Bull. 36 (2), 314–320. Francis, A., Fink, M., Appiani, F., Bertelsen, A., Bolwig, T.G., Braunig, P., Caroff, S.N., Carroll, B.T., Cavanna, A.E., Cohen, D., Cottencin, O., Cuesta, M.J., Daniels, J., Dhossche, D., Fricchione, G.L., Gazdag, G., Ghaziuddin, N., Healy, D., Klein, D., Kruger, S., Lee, J.W., Mann, S.C., Mazurek, M., McCall, W.V., McDaniel, W.W., Northoff, G., Peralta, V., Petrides, G., Rosebush, P., Rummans, T.A., Shorter, E., Suzuki, K., Thomas, P., Vaiva, G., Wachtel, L., 2010. Catatonia in diagnostic and statistical manual of mental disorders, fifth edition. J. ECT 26 (4), 246–247. Gorsuch, R.L., 1993. Factor Analysis 2ed. Lawrence Erlbaum Associates, New Jersey. Heckers, S., Tandon, R., Bustillo, J., 2010. Catatonia in the DSM—shall we move or not? Schizophr. Bull. 36 (2), 205–207. Kahlbaum, K., 1874. Die Katatonie oder das Spannungsirresein. Hirschwald, Berlin. Kraepelin, E., 1919. Dementia Praecox and Paraphrenia. Chicago Medical Book Co, Chicago. Kruger, S., Bagby, R.M., Hoffler, J., Braunig, P., 2003. Factor analysis of the catatonia rating scale and catatonic symptom distribution across four diagnostic groups. Compr. Psychiatry 44 (6), 472–482. Kruskal, J.B., Wish, M., 1978. Multidimensional Scaling. Sage, Newbury, CA. Levine, S.Z., Rabinowitz, J., Rizopoulos, D., 2011. Recommendations to improve the positive and negative syndrome scale (PANSS) based on item response theory. Psychiatry Res. 188 (3), 446–452.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
J.E. Wilson et al. / Schizophrenia Research xxx (2015) xxx–xxx Lopez-Canino, A., F.A., 2004. Drug-induced catatonia. In: Caroff, S.N., M.S., Francis, A., Fricchione, G.L. (Eds.), Catatonia: From Psychopathology to Neurobiology. American Psychiatric Press, Inc., Washington, D.C., pp. 129–139. McKenna, P.J., Lund, C.E., Mortimer, A.M., Biggins, C.A., 1991. Motor, volitional and behavioural disorders in schizophrenia. 2: the ‘conflict of paradigms’ hypothesis. Br. J. Psychiatry 158, 328–336. Morrison, J.R., 1973. Catatonia. Retarded and excited types. Arch. Gen. Psychiatry 28 (1), 39–41. Morrison, J.R., 1974a. Catatonia: prediction of outcome. Compr. Psychiatry 15 (4), 317–324. Morrison, J.R., 1974b. Karl Kahlbaum and catatonia. Compr. Psychiatry 15 (4), 315–316. Morrison, J.R., 1975. Catatonia: diagnosis and management. Hosp. Community Psychiatry 26 (2), 91–94. Northoff, G., Koch, A., Wenke, J., Eckert, J., Boker, H., Pflug, B., Bogerts, B., 1999. Catatonia as a psychomotor syndrome: a rating scale and extrapyramidal motor symptoms. Mov. Disord. 14 (3), 404–416. Oulis, P., Lykouras, L., Tomaras, V., Panayotopoulou, V., Gournellis, R., Stefanis, C., 1997. DSM-IV catatonic features among psychiatric inpatients: a preliminary study. Eur. Psychiatry 12 (8), 412–414. Peralta, V., Cuesta, M.J., 2001. Motor features in psychotic disorders. I. Factor structure and clinical correlates. Schizophr. Res. 47 (2–3), 107–116. Peralta, V., Cuesta, M.J., Mata, I., Serrano, J.F., Perez-Nievas, F., Natividad, M.C., 1999. Serum iron in catatonic and noncatatonic psychotic patients. Biol. Psychiatry 45 (6), 788–790. Peralta, V., Campos, M.S., de Jalon, E.G., Cuesta, M.J., 2010. DSM-IV catatonia signs and criteria in first-episode, drug-naive, psychotic patients: psychometric validity and response to antipsychotic medication. Schizophr. Res. 118 (1–3), 168–175.
7
Rizopoulos, D., 2006. ltm: an R package for latent variable modelling and item response theory analyses. J. Stat. Softw. 17 (5), 1–25. Rosebush, P.I., Mazurek, M.F., 2010. Catatonia and its treatment. Schizophr. Bull. 36 (2), 239–242. Sienaert, P., Rooseleer, J., De Fruyt, J., 2011. Measuring catatonia: a systematic review of rating scales. J. Affect. Disord. 135 (1–3), 1–9. Starkstein, S.E., Petracca, G., Teson, A., Chemerinski, E., Merello, M., Migliorelli, R., Leiguarda, R., 1996. Catatonia in depression: prevalence, clinical correlates, and validation of a scale. J. Neurol. Neurosurg. Psychiatry 60 (3), 326–332. Tabachnick, B.G., Fidell, L.S., 2001. Using Multivariate Statistics. 4 ed. Harper & Colins, NY. Tandon, R., Heckers, S., Bustillo, J., Barch, D.M., Gaebel, W., Gur, R.E., Malaspina, D., Owen, M.J., Schultz, S., Tsuang, M., van Os, J., Carpenter, W., 2013. Catatonia in DSM-5. Schizophr. Res. 150, 26–30. Taylor, M.A., Abrams, R., 1977. Catatonia. Prevalence and importance in the manic phase of manic-depressive illness. Arch. Gen. Psychiatry 34 (10), 1223–1225. Taylor, M.A., Fink, M., 2003. Catatonia in psychiatric classification: a home of its own. Am. J. Psychiatry 160 (7), 1233–1241. Ungvari, G.S., Goggins, W., Leung, S.K., Gerevich, J., 2007. Schizophrenia with prominent catatonic features (‘catatonic schizophrenia’). II. Factor analysis of the catatonic syndrome. Prog. Neuropsychopharmacol Biol. Psychiatry 31 (2), 462–468. Vesperini, S., Papetti, F., Pringuey, D., 2010. Are catatonia and neuroleptic malignant syndrome related conditions? Encéphale 36 (2), 105–110. Wong, E., Ungvari, G.S., Leung, S.K., Tang, W.K., 2007. Rating catatonia in patients with chronic schizophrenia: Rasch analysis of the Bush-Francis Catatonia Rating Scale. Int. J. Methods Psychiatr. Res. 16 (3), 161–170.
Please cite this article as: Wilson, J.E., et al., The diagnostic criteria and structure of catatonia, Schizophr. Res. (2015), http://dx.doi.org/10.1016/ j.schres.2014.12.036
