- Email: [email protected]
Comparisons of methamphetamine psychotic and schizophrenic symptoms: A differential item functioning analysis
Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
Contents lists available at ScienceDirect
Progress in Neuro-Psychopharmacology & Biological Psychiatry j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p n p
Comparisons of methamphetamine psychotic and schizophrenic symptoms: A differential item functioning analysis Manit Srisurapanont a,⁎, Suwanna Arunpongpaisal b, Kiyoshi Wada c, John Marsden d, Robert Ali e, Ronnachai Kongsakon f a
Department of Psychiatry, Chiang Mai University, Chiang Mai, Thailand Department of Psychiatry, Khon Kaen University, Khon Kaen, Thailand Division of Drug Dependence Research, National Institute of Mental Health, Chiba-ken, Japan d National Addiction Centre, Institute of Psychiatry, University of London, London, UK e Department of Clinical and Experimental Pharmacology, University of Adelaide, Adelaide, Australia f Department of Psychiatry, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand b c
a r t i c l e
i n f o
Article history: Received 15 November 2010 Received in revised form 2 January 2011 Accepted 15 January 2011 Available online 26 January 2011 Keywords: Differential item functioning Methamphetamine Psychosis Schizophrenia Symptom
a b s t r a c t The concept of negative symptoms in methamphetamine (MA) psychosis (e.g., poverty of speech, flatten affect, and loss of drive) is still uncertain. This study aimed to use differential item functioning (DIF) statistical techniques to differentiate the severity of psychotic symptoms between MA psychotic and schizophrenic patients. Data of MA psychotic and schizophrenic patients were those of the participants in the WHO MultiSite Project on Methamphetamine-Induced Psychosis (or WHO-MAIP study) and the Risperidone Long-Acting Injection in Thai Schizophrenic Patients (or RLAI-Thai study), respectively. To confirm the unidimensionality of psychotic syndromes, we applied the exploratory and confirmatory factor analyses (EFA and CFA) on the eight items of Manchester scale. We conducted the DIF analysis of psychotic symptoms observed in both groups by using nonparametric kernel-smoothing techniques of item response theory. A DIF composite index of 0.30 or greater indicated the difference of symptom severity. The analyses included the data of 168 MA psychotic participants and the baseline data of 169 schizophrenic patients. For both data sets, the EFA and CFA suggested a three-factor model of the psychotic symptoms, including negative syndrome (poverty of speech, psychomotor retardation and flatten/incongruous affect), positive syndrome (delusions, hallucinations and incoherent speech) and anxiety/depression syndrome (anxiety and depression). The DIF composite indexes comparing the severity differences of all eight psychotic symptoms were lower than 0.3. The results suggest that, at the same level of syndrome severity (i.e., negative, positive, and anxiety/depression syndromes), the severity of psychotic symptoms, including the negative ones, observed in MA psychotic and schizophrenic patients are almost the same. © 2011 Elsevier Inc. All rights reserved.
1. Introduction Methamphetamine (MA) psychosis is a state of MA intoxication with psychotic symptoms, commonly presented with delusions and hallucinations. This psychotic condition has been considered as a common and serious consequence of chronic, high dose, and/or continuous use of MA (Griffith et al., 1972). It is commonly described as closely simulating paranoid schizophrenia (Bell, 1965).
Abbreviations: CFA, confirmatory factor analysis; DIF, differential item functioning; EFA, exploratory factor analysis; IRT, item response theory; MA, methamphetamine; RLAI-Thai, Risperidone Long-Acting Injection in Thai Schizophrenic Patients; WHOMAIP, WHO Multi-Site Project on Methamphetamine-Induced Psychosis. ⁎ Corresponding author at: Department of Psychiatry, Faculty of Medicine, Chiang Mai University, Muang, Chiang Mai 50200, Thailand. Tel.: +66 53 945422; fax: +66 53 945426. E-mail address: [email protected] (M. Srisurapanont). 0278-5846/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.pnpbp.2011.01.014
The similarities in many aspects of MA psychosis and schizophrenia have made amphetamine a primary psychotomimetic model agent in schizophrenia research. It is likely that the resemblance is caused by the altered function of mesolimbic dopamine systems and prefrontal cortical function (Robinson and Becker, 1986). Although clinical research is necessary for the development of MA psychosis services, few studies in this area have been carried out. Of many respects, symptom studies may be a priority area because these results are basic knowledge for further studies of MA psychosis, e.g., etiology, course, prognosis, and treatment. So far, most of the findings on the resemblance between MA psychosis and schizophrenia have mainly focused on positive psychotic symptoms, in particular delusions and hallucinations. The concept of negative symptoms in MA psychosis (e.g., poverty of speech, flatten affect, and loss of drive) is not new but still uncertain. Blunted affect and diminished spontaneity accompanied with paranoid-hallucinatory state were described in 1960s (Yui et al.,
960
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
2000). In a study of 11 MA psychotic inpatients assessed by using the Scale for the Assessment of Negative Symptoms, Japanese investigators found that most patients showed a significant impairment on avolition-apathy, anhedonia-asociality, and attentional impairment (Tomiyama, 1990). Srisurapanont et al. (2003) found negative psychotic symptoms in 21.4% of 168 MA psychotic patients with various ethnicity (Srisurapanont et al., 2003) In this later study, the results of an exploratory factor analysis (EFA) also showed an independent syndrome of negative symptoms with a variance of 26.6%. Despite the above-mentioned evidence, negative symptoms of MA psychosis are still viewed as less severe and/or prevalent (Zorick et al., 2008). Differential item functioning (DIF) statistical techniques are based on the principle that if different groups of patients (e.g., males vs. females) have the same level of disease severity (e.g., depression, psychosis), they should be rated or response similarly on an individual rating item of a measure, regardless of group membership. DIF assessment proceeds by controlling for an estimate of the underlying construct and then examining whether individuals in different groups have a similar distribution of responses to a particular item. The statistical techniques that directly compare item severity between groups are inferior to DIF analyses by the fact that the difference found by a direct comparison may not be a true phenomenological difference. It may be just reflective of greater overall symptom severity in one group versus another. In contrast, the existence of DIF between different groups on a symptom indicates the real difference of symptom severity because the overall disease or syndrome severity has been controlled. Recently, we have finished two studies: (i) the WHO Multi-Site Project on Methamphetamine-Induced Psychosis (WHO-MAIP study), a cross-sectional observation study aiming to evaluate MA psychosis (Ali et al., 2006, Ali et al., 2010, Srisurapanont, et al., 2003), and (ii) Risperidone Long-Acting Injection in Thai Schizophrenic Patients (RLAI-Thai study), a three-month, non-randomized, open-label, single-arm study of risperidone long-acting injection in Thai individuals with schizophrenia (Arunpongpaisal et al., 2010). The psychotic symptoms of participants in both studies were assessed using the Manchester scale (Krawiecka et al., 1977). Because the symptoms of these two studies were comparable, we proposed to apply the DIF statistical techniques to differentiate the severity of psychotic symptoms between these populations. 2. Methods Data of MA psychotic and schizophrenic patients were those of the participants in the WHO-MAIP and the RLAI-Thai studies, respectively. Both studies were carried out in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans. The protocol of each study was approved by the Institution Review Board or the Ethic Committee responsible for the participating hospital or institution. Written informed consent was obtained from each participant after the study details had been fully explained. 2.1. Participants The participants of WHO-MAIP study were MA psychotic patients admitted in many hospitals located in Australia, Japan, the Philippines, and Thailand. In addition to the evidence of MA use during the week prior to the admission, the diagnosis of MA psychosis was confirmed using the Mini-International Neuropsychiatric Interview-Plus (MINIPlus), Module M. The MINI-Plus is a more detailed version of the MINI, a structured clinical interview for the DSM-III-R diagnosis (Sheehan et al., 1998). It includes additional questions to ascertain the relationship between a psychiatric condition and a physical illness
or drug use. These patients were assessed only once within 3–7 days after the admission. The participants of RLAI-Thai study were Thai inpatients and outpatients with DSM-IV schizophrenia, who did not fully respond or could not tolerate their antipsychotic medications. They were willing to switch their previous antipsychotic medications to 25, 37.5, or 50 mg of RLAI given by intramuscular injection. Exclusion criteria included (i) serious unstable medical condition, including recent and present clinically relevant laboratory abnormalities; (ii) history of alcohol dependence or drug abuse (except nicotine) within 3 months of entry into the trial, (iii) on clozapine during the last 3 months; (iv) history or current symptoms of tardive dyskinesia; (v) history of neuroleptic malignant syndrome; (vi) pregnant or breast-feeding female; (vii) female patient of childbearing potential without adequate contraception; (viii) participation in an investigational drug trial in the 30 days prior to selection; and (ix) known intolerance/non-responder to risperidone. Other psychiatric comorbidity was not assessed. Therefore, the patients with comorbidity of other psychiatric illnesses were allowed to participate in this study. Participants were assessed prior to RLAI administration (baseline) and then again after 3 months and 6 months of monthly RLAI treatment. Because the psychotic symptoms at baseline were the most severe ones, the data at this time point were applied in this study. 2.2. Measures In WHO-MAIP study, the participants' psychotic symptoms were assessed using the Manchester scale. This 5-point scale (0 = absent, 1 = mild, 2 = moderate, 3 = marked and 4 = severe) assesses eight psychiatric symptoms commonly found in psychotic patients, including depression, anxiety, delusions, hallucinations, incoherence speech, poverty of speech, flatten/incongruous affect, and psychomotor retardation. For each item of the scale, its inter-rater reliability (Kandall's coefficient of concordance W) is between 0.6 and 0.9 (Krawiecka, et al., 1977). Its concurrent validity has been tested with the Schedules for the Assessment of Positive Symptoms (SAPS) (Andreasen, 1984) and the Schedules for the Assessment of Negative Symptoms (SANS) (Andreasen, 1983). While the Manchester scale items of delusions, hallucinations, and incoherence are significantly correlated with the total SAPS score (r's = 0.53–0.64), its items of poverty of speech, affect flattening, and psychomotor retardation are also significantly correlated with the total SANS score (r's = 0.62–0.85) (Jackson et al., 1990). Except disorganized and catatonic behavior, the Manchester scale covers all DSM-IV schizophrenic symptoms and proposed remission criteria for schizophrenia (American Psychiatric Association, 2000, Andreasen et al., 2005). The modified version of Manchester scale was applied in the RLAIThai study. As used in an antipsychotic trial (Johnstone et al., 1978), the item of flatten/incongruous affect was separated into two items of flatten and incongruous affects. To make the scores of these two items comparable to the single one of flatten/incongruous affect in the WHO-MAIP study, the highest score of these two items in each participant was used as the representative of flatten/incongruous affect item. The different versions of Manchester scale used in both studies were caused by the decision of each investigator team. Both versions were administered by certified psychiatrists, who were well-trained on the use of Manchester scale. 2.3. Data analyses 2.3.1. Unidimensionality The DIF is typically examined on the basis that the set of items is intended to measure a single underlying construct (i.e., the unidimension of items). An examination of the underlying dimension of an item set is therefore the first step in DIF detection. More
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
961
Table 1 Comparisons of demographic data and severity of symptoms between 168 methamphetamine psychotic patients and 169 schizophrenic patients.a Significant differencea
Methamphetamine psychotic patients (n = 168)
Schizophrenic patients (n = 169)
n
n
Sex (male,%)
127 (75.6%) Mean (SD)
114 (67.5%) Mean (SD)
χ2 = 2.74, df = 1, p = 0.12 Mean (SD)
Age Age at first onset of psychotic symptoms Total Manchester scale score Negative syndrome Poverty of speech Psychomotor retardation Flatten/incongruouse affect Positive syndrome Delusions Hallucinations Incoherence speech Anxiety/depression syndrome Depression Anxiety
27.1 25.0 10.7 2.7 0.7 0.8 1.2 5.3 2.1 2.0 1.1 2.8 1.1 1.7
29.8 (14.0) 23.2 (8.4) 8.0 (5.5) 2.9 (2.5) 0.7 (0.9) 0.9 (1.0) 1.3 (1.0) 3.7 (3.0) 1.4 (1.3) 1.4 (1.3) 0.9 (1.0) 1.5 (1.7) 0.5 (0.8) 1.0 (1.1)
t = 2.20, p = 0.03 t = 1.89, p = 0.06 Z = −4.77, p b 0.01 Z = −0.54, p = 0.59 Z = −0.18, p = 0.86 Z = −0.40, p = 0.69 Z = −1.36, p = 0.17 Z = −4.86, p b 0.01 Z = −5.10, p b 0.01 Z = −4.03, p b 0.01 Z = −2.11, p = 0.04 Z = −6.89, p b 0.01 Z = −6.33, p b 0.01 Z = −5.70, p b 0.01
a
(7.6) (9.1) (4.9) (2.3) (0.9) (0.8) (1.0) (2.9) (1.2) (1.4) (1.1) (1.7) (1.0) (1.0)
Except a Chi-square test for sex and a Student's t test for age items, Mann–Whitney U tests were used for all comparisons.
specifically, ones should examine dimensionality separately for each group (i.e., MP or schizophrenic group in the present analyses) by determining if the same number of factors is found in each group (Teresi and Fleishman, 2007). In addition, confirmatory factor analysis (CFA) should be applied to test the single-factor model fit of each item set. The EFA on the eight items of Manchester scale scores obtained from the MA psychotic group was completed in our previous study (Srisurapanont et al., 2003). The same methods were applied on the scores obtained from schizophrenic patients. All eight items were subjected to principal-component analysis for identifying the distinct factors. Eigenvalue one test was used to keep or discard factors. Finally, varimax rotation was performed to elicit the factor components. The EFA was examined by using the SPSS 17.0. After the combination of both data sets, each factor found on the EFA was further confirmed by using the chi-square test of a single-factor model performed by the AMOS 18.0. 2.3.2. DIF analyses We employed the nonparametric kernel-smoothing techniques of item response theory (IRT) implemented in the software (TestGraf) developed by Ramsay (1991, 2001). The strengths of these techniques are that non-parametric IRT models do not require complex
estimation procedures, can be applied to relatively small data sets, are less imposing concerning distributional form of item response functions, and help to avoid misleading results obtained from parametric IRT models. In the present analyses, we determined the item response functions directly from the data without forcing the data to conform to a logistic IRT model. TestGraf analyzes the severity of symptoms observed in the participants of both groups. A response characteristic curve shows how the item score varies as a function of what the test is designed to measure, that is, its latent trait (i.e., severity of each syndrome in the present analyses). To examine DIF, this software calculates a weighted average of the squared difference between the focal group's probability of endorsing an item and the reference group's probability of endorsing an item. Thus, a composite index of DIF is obtained after the comparison of response characteristic curves. For each item, there are two curves of both groups being compared. The curves are numbered in the order in which the files of each group have been selected. The first group is called the focal group (i.e., MA psychotic patients), and the other group is called a reference group (i.e., schizophrenic patients). For each focal group, a summary measure of how far the curve this group is from that of the reference group is calculated. The extent to which the curves differ is called differential item functioning (DIF). If this measure is
Table 2 Exploratory factor analysis of eight-item Manchester scale in 168 methamphetamine psychotic and 169 schizophrenic patients. Methamphetamine psychotic patients (n = 168)a
Schizophrenic patients (n = 169)
Factor 1
Factor 1
Factor 2
Factor 3
Factor 2
Factor 3
0.89 0.88 0.75
-0.17 0.23 0.20
0.10 -0.01 0.39
0.13 0.17 -0.29
-0.01 0.07 0.33
0.56 0.32 -2.16
0.73 0.79 0.77
-0.11 0.27
0.85 0.76
0.10 0.19
0.85 0.85
0.08 0.14
1.92 23.95
1.51 18.81
2.27 28.35
2.00 24.93
1.94 24.24
Negative syndrome Poverty of speech Psychomotor retardation Flatten/incongruous affect
0.86 0.83 0.75
0.07 0.04 0.28
0.02 0.25 0.03
Positive syndrome Delusions Hallucinations Incoherent speech
0.08 0.07 0.32
0.82 0.79 0.67
Anxiety/depression syndrome Depression Anxiety
0.19 0.05 2.13 26.64
Eigenvalue Per cent variance a
(Srisurapanont et al., 2003, reprinted with permission).
962
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
statistically significant, the focal and reference groups are not rated in the same way on that item. In the present analyses, a composite index (measure) of DIF for each focal group was calculated. Taking account the Santor and colleagues' criteria, a value of 0.30 or greater was considered as the existence of DIF, where a value of 0.10 or less suggested little or no DIF (Santor et al., 1994).
A
3. Results 3.1. Characteristics of the participants Of 181 patients participating in the WHO-MAIP study, the Manchester scale scores were completed in 168 participants: 32 participants from Australia, 36 from Japan, 50 from the Philippines, and 50 from Thailand. Their means (SDs) age at first MA use and duration of MA regular use were 19.7 (5.9) years and 3.8 (5.4) years, respectively. The mean (SD) of the maximum amount of MA use during the 12 months prior to the assessment was 26.8 (130.9) g. The data on number of treatment for MA use were available in 140 patients. Of these, 70 patients (50%) had ever received the treatment with a mean (SD) of 1.2 (2.1) times. Of 184 Thai schizophrenic and schizoaffective patients participated in the RLAI-Thai study, nine patients were excluded due to their diagnosis of schizoaffective disorder and the other six participants with schizophrenia were excluded due to the incompletion of Manchester scale scores. Therefore, the data of 169 schizophrenic participants in the RLAI-Thai study were included in the present analyses. Table 1 shows the demographic data and symptom severity of both groups. Of note, MA psychotic group were significantly younger (p = 0.03) but tended to have a later onset of psychotic symptoms (p = 0.06). In addition, their positive and anxiety/depressive syndromes, as well as symptoms, were more severe than those of schizophrenic patients (p's b 0.01 for all comparisons).
B
C
3.2. Factor structure of psychotic symptoms Table 2 shows the previous EFA results of 168 MA psychotic patients and the present results of 169 schizophrenic patients. Although the item of delusions in schizophrenic patients moderately loaded (0.56) on anxiety/depression syndrome, all highest factor loadings of each item were located on the factors as found in the MA psychotic patients. These results suggested a three-factor model of psychotic symptoms in both groups, including a negative syndrome (poverty of speech, psychomotor retardation and flatten/incongruous affect), a positive syndrome (delusions, hallucinations and incoherent speech), and an anxiety/depression syndrome (anxiety and depression). The following chi-square test of single-factor model fit also confirmed the goodness-of-fit of each factor. 3.3. Differential item functioning analysis Figs. 1A–1C, 2A–2C, and 3A and 3B show the response characteristic curves of each item generated from the Manchester scale scores obtained from the MA psychotic and schizophrenic groups. All eight pairs of the curves were almost parallel with all DIF composite indexes lower than 0.3. Except the minimal DIF of incoherent speech (a DIF composite index of 0.142), the composite indexes of the other seven symptoms were lower than 0.1. 4. Discussion 4.1. Main findings The DIF composite indexes of all eight items that were lower than 0.3 indicate the similarity of psychotic symptoms found in MA
Fig. 1. Differential item functioning of negative symptoms: 168 methamphetamine psychotic patients (group 1) vs. 169 schizophrenic patients (group 2) (X axis shows the combined scores of 3 negative symptoms). (Y axis shows the scores of each item). (A) Poverty of speech. (B) Psychomotor retardation. (C) Flatten and incongruous affect.
psychotic and schizophrenic patients. The composite index being lower than 0.1 in the comparisons of seven symptoms suggests that, except the incoherence speech, these seven symptoms, including the negative ones, are rated almost the same in both groups. It is noteworthy that the direct comparison of negative-symptom severity by using a simple statistical technique of Mann–Whitney U test also yielded the same results of nonsignificant differences between groups. Although the direct comparisons of syndrome and symptom severity by using Mann–Whitney U test revealed that MA psychotic
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
A
A
B
B
C
963
Fig. 3. Differential item functioning of anxiety/depression symptoms: 168 methamphetamine psychotic patients (group 1) vs. 169 schizophrenic patients (group 2) (X axis shows the combined scores of 2 anxiety/depression symptoms). (Y axis shows the scores of each item). (A) Depression. (B) Anxiety.
4.2. Relevance of present and previous findings
Fig. 2. Differential item functioning of positive symptoms: 168 methamphetamine psychotic patients (group 1) vs. 169 schizophrenic patients (group 2) (X axis shows the combined scores of 3 positive symptoms). (Y axis shows the scores of each item). (A) Delusions. (B) Hallucinations. (C) Incoherence speech.
patients had more severe positive symptoms (i.e., delusions, hallucinations, incoherence speech), depression, and anxiety (see Table 1), these findings may not be true phenomenological differences. They may be caused by the greater severity of overall positive and anxiety/depression syndromes. This issue is similar to the problems found in many studies directly comparing the severity of bipolar and major depressive symptoms in which Weinstock and colleagues have criticized that the differential symptom expressions reported in those studies may be a reflection of greater overall symptom severity in one group versus another (Weinstock et al., 2010).
The factor structure of MA psychotic and schizophrenic patients found in the present analysis, especially the negative factor, is similar to that of a previous study (Harvey et al., 1996). By using five psychotic symptoms of the Manchester scale, Harvey and colleagues conducted a factor analysis of psychotic symptoms in three groups of schizophrenic patients, including those living in community (n = 320), acute inpatients (n = 49), and long-stay inpatients (n = 35). Their study found that poverty of speech, psychomotor retardation, and flatten/incongruous affect highly loaded in the negative factor, and delusions and hallucination highly loaded in the same factor. To our knowledge, this is the first DIF analysis comparing MA psychotic and schizophrenic symptoms. These findings extend the previous knowledge on prevalence and factor structure of negative symptoms in MA psychotic patients (Srisurapanont et al., 2003; Tomiyama, 1990). The present findings suggest that negative symptoms are not only common in MA psychotic patients but also rated exactly in the same way as those found in schizophrenic patients. In addition, the positive and anxiety/depression symptoms observed in both populations are also similar. 4.3. Limitation of the present analysis There are some limitations of the present study. First, the sample size of 168 or 169 for each group is relatively small for the current concept of DIF analysis. The minimum sample size for logistic regression DIF analysis, which is a statistic technique similar to the
964
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
one used in TestGraf, is usually in the range of 100–200 per group (Lai et al., 2005). However, the results of a recent study suggested that a minimum of 200 respondents per group may be needed to ensure the adequate performance (N80% power) (Scott et al., 2009). Second, it is impossible to exclude the effects of antipsychotic medications on the findings of this study. According to the nature of disease, almost all participants in RLAI-Thai study had been taking antipsychotic medications for months or years before the assessment. In the contrary, as patients with short-term psychotic symptoms, many participants in the WHO-MAIP study had been taking antipsychotic medications for only days or weeks. Moreover, a few of them were not taking an antipsychotic medication at all. Third, because the schizophrenic patients in this study were those not responding well or not being able to tolerate some antipsychotic medications, they might not represent the general schizophrenic patients. Fourth, the small number of items included in the Manchester scale may not cover some psychotic symptoms currently described in schizophrenic patients, e.g., disorganized or catatonic behavior. Last, the ethnicities of participants in both groups are not the same. While the MA psychotic groups composed of Australians, Japanese, the Filipinos and Thai, only Thai patients participated in the RLAI-Thai study. Whether the ethnicity difference had an impact on the expression of psychotic symptoms remains unknown. 4.4. Implications of the findings Despite the above-mentioned limitations, the findings that psychotic symptoms in MA psychotic and schizophrenic patients are almost the same may suggest a share of pathophysiology of MA psychosis and schizophrenia. The present study's results, therefore, support a recent proposal of using amphetamine psychosis as a human model for studying the onset and course of schizophrenia (Hermens et al., 2009). Although this human model is still underutilized, the concept of using amphetamine as an environmental pathogen may be another strategy for the study of schizophrenia. 4.5. Conclusion The psychotic symptoms, including the negative ones, found in MA psychotic and schizophrenic patients are almost the same. More specifically, at the same level of syndrome severity (i.e., negative, positive, and anxiety/depression syndromes), the severity of each symptom observed in both groups are not much different. Further studies in large sample sizes are warranted. Conflict of interest None. Acknowledgments We wish to thank coauthors of the WHO-MAIP and the RLAI-Thai studies who allowed us to reanalyze the data.
References Ali R, Baigent M, Marsden J, Montiero M, Srisurapanont M, Sunga A, et al. WHO MultiSite Project on Methamphetamine-induced Psychosis: A descriptive report of findings from participating countries. Adelaide (South Australia): Drug & Alcohol Services South Australia; 2006. Ali R, Marsden J, Srisurapanont M, Sunga A, Baigent M, Monteiro M. Methamphetamine psychosis in Australia, Philippines, and Thailand: Recommendations for acute care and clinical inpatient management. Addict Disord Their Treat 2010;9:143–9. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4-TR ed. Washington (DC): American Psychiatric Association; 2000. Andreasen NC. The scale for the assessment of negative symptoms (SANS). Iowa City: University of Iowa; 1983. Andreasen NC. The scale for the assessment of positive symptoms (SAPS). Iowa City: University of Iowa; 1984. Andreasen NC, Carpenter JWT, Kane JM, Lasser RA, Marder SR, Weinberger DR. Remission in schizophrenia: proposed criteria and rationale for consensus. Am J Psychiatry 2005;162:441–9. Arunpongpaisal S, Srisurapanont M, Kongsakon R, Kitiwattanagul K, Samanwongthai U. Risperidone long-acting injection (RLAI): the 12-week efficacy and tolerability in Thai patients with chronic schizophrenia. J Med Assoc Thai 2010;93:343–50. Bell DS. Comparison of amphetamine psychosis and schizophrenia. Br J Psychiatry 1965;111:701–7. Griffith JD, Cavanaugh J, Held J, Oates J. Dextroamphetamine: evaluation of psychomimetic properties in man. Arch Gen Psychiatry 1972;26:97-100. Harvey C, Curson D, Pantelis C, Taylor J, Barnes T. Four behavioural syndromes of schizophrenia. Br J Psychiatry 1996;168:562–70. Hermens DF, Lubman DI, Ward PB, Naismith SL, Hickie IB. Amphetamine psychosis: a model for studying the onset and course of psychosis. MJA 2009;190:S22–5. Jackson HJ, Burgess PM, Minas IH, Joshua SD. Psychometric properties of the Manchester scale. Acta Psychiatr Scand 1990;81:108–13. Johnstone EC, Crow TJ, Frith CD, Carney MW, Price JS. Mechanism of the antipsychotic effect in the treatment of acute schizophrenia. Lancet 1978;1(8069):848–51. Krawiecka M, Goldberg D, Vaughan M. A standardized psychiatric assessment scale for rating chronic psychotic patients. Acta Psychiatr Scand 1977;55:299–308. Lai JS, Teresi J, Gershon R. Procedures for the analysis of differential item functioning (DIF) for small sample size. Eval Health Prof 2005;28:283–94. Ramsay JO. Kernel smoothing approaches to non parametric item curve estimation. Psychometrika 1991;56:611–30. Ramsay JO. TesGraf: A program for the graphical analysis of multiple choice test and questionnaire data. Montreal, Canada: McGill University; 2001. Robinson T, Becker JB. Enduring changes in brain and behavior produced by chronic amphetamine administration: a review and evaluation of animal models of amphetamine psychosis. Brain Res 1986;396:157–98. Santor DA, Ramsay JO, Suroff DC. Nonparametric item analyses of the Beck Depression Inventory: evaluating gender item bias and response option weights. Psychol Assess 1994;6:255–70. Scott NW, Fayers PM, Aaronson NK, Bottomley A, de Graeff A, Groenvold M, et al. A simulation study provided sample size guidance for differential item functioning (DIF) studies using short scales. J Clin Epidemiol 2009;62:288–95. Sheehan D, Lecrubier Y, Sheehan K, Amorim P, Janavs J, Weiller E, et al. The MiniInternational Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59:22–33. Srisurapanont M, Ali R, Marsden J, Sunga A, Wada K, Montiero M. Psychotic symptoms in methamphetamine psychotic in-patients. Int J Neuropsychopharmacol 2003;6:347–52. Teresi JA, Fleishman JA. Differential item functioing and health assessment. Qual Life Res 2007;16:33–42. Tomiyama G. Chronic schizophrenia-like states in methamphetamine psychosis. Jpn J Psychiatry Neurol 1990;44:531–9. Weinstock LM, Strong D, Uebelacker LA, Miller IW. DSM-IV depressive symptom expression among individuals with a history of hypomania: a comparison to those with or without a history of mania. J Psychiatr Res 2010;44:979–85. Yui K, Ikemoto S, Ishiguro T, Goto K. Studies of amphetamine or methamphetamine psychosis in Japan. Ann NY Acad Sci 2000;914:1-12. Zorick TS, Rad D, Rim C, Tsuang J. An overview of methamphetamine-induced psychotic syndromes. Addict Disord Their Treat 2008;7:143–56.
Contents lists available at ScienceDirect
Progress in Neuro-Psychopharmacology & Biological Psychiatry j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p n p
Comparisons of methamphetamine psychotic and schizophrenic symptoms: A differential item functioning analysis Manit Srisurapanont a,⁎, Suwanna Arunpongpaisal b, Kiyoshi Wada c, John Marsden d, Robert Ali e, Ronnachai Kongsakon f a
Department of Psychiatry, Chiang Mai University, Chiang Mai, Thailand Department of Psychiatry, Khon Kaen University, Khon Kaen, Thailand Division of Drug Dependence Research, National Institute of Mental Health, Chiba-ken, Japan d National Addiction Centre, Institute of Psychiatry, University of London, London, UK e Department of Clinical and Experimental Pharmacology, University of Adelaide, Adelaide, Australia f Department of Psychiatry, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand b c
a r t i c l e
i n f o
Article history: Received 15 November 2010 Received in revised form 2 January 2011 Accepted 15 January 2011 Available online 26 January 2011 Keywords: Differential item functioning Methamphetamine Psychosis Schizophrenia Symptom
a b s t r a c t The concept of negative symptoms in methamphetamine (MA) psychosis (e.g., poverty of speech, flatten affect, and loss of drive) is still uncertain. This study aimed to use differential item functioning (DIF) statistical techniques to differentiate the severity of psychotic symptoms between MA psychotic and schizophrenic patients. Data of MA psychotic and schizophrenic patients were those of the participants in the WHO MultiSite Project on Methamphetamine-Induced Psychosis (or WHO-MAIP study) and the Risperidone Long-Acting Injection in Thai Schizophrenic Patients (or RLAI-Thai study), respectively. To confirm the unidimensionality of psychotic syndromes, we applied the exploratory and confirmatory factor analyses (EFA and CFA) on the eight items of Manchester scale. We conducted the DIF analysis of psychotic symptoms observed in both groups by using nonparametric kernel-smoothing techniques of item response theory. A DIF composite index of 0.30 or greater indicated the difference of symptom severity. The analyses included the data of 168 MA psychotic participants and the baseline data of 169 schizophrenic patients. For both data sets, the EFA and CFA suggested a three-factor model of the psychotic symptoms, including negative syndrome (poverty of speech, psychomotor retardation and flatten/incongruous affect), positive syndrome (delusions, hallucinations and incoherent speech) and anxiety/depression syndrome (anxiety and depression). The DIF composite indexes comparing the severity differences of all eight psychotic symptoms were lower than 0.3. The results suggest that, at the same level of syndrome severity (i.e., negative, positive, and anxiety/depression syndromes), the severity of psychotic symptoms, including the negative ones, observed in MA psychotic and schizophrenic patients are almost the same. © 2011 Elsevier Inc. All rights reserved.
1. Introduction Methamphetamine (MA) psychosis is a state of MA intoxication with psychotic symptoms, commonly presented with delusions and hallucinations. This psychotic condition has been considered as a common and serious consequence of chronic, high dose, and/or continuous use of MA (Griffith et al., 1972). It is commonly described as closely simulating paranoid schizophrenia (Bell, 1965).
Abbreviations: CFA, confirmatory factor analysis; DIF, differential item functioning; EFA, exploratory factor analysis; IRT, item response theory; MA, methamphetamine; RLAI-Thai, Risperidone Long-Acting Injection in Thai Schizophrenic Patients; WHOMAIP, WHO Multi-Site Project on Methamphetamine-Induced Psychosis. ⁎ Corresponding author at: Department of Psychiatry, Faculty of Medicine, Chiang Mai University, Muang, Chiang Mai 50200, Thailand. Tel.: +66 53 945422; fax: +66 53 945426. E-mail address: [email protected] (M. Srisurapanont). 0278-5846/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.pnpbp.2011.01.014
The similarities in many aspects of MA psychosis and schizophrenia have made amphetamine a primary psychotomimetic model agent in schizophrenia research. It is likely that the resemblance is caused by the altered function of mesolimbic dopamine systems and prefrontal cortical function (Robinson and Becker, 1986). Although clinical research is necessary for the development of MA psychosis services, few studies in this area have been carried out. Of many respects, symptom studies may be a priority area because these results are basic knowledge for further studies of MA psychosis, e.g., etiology, course, prognosis, and treatment. So far, most of the findings on the resemblance between MA psychosis and schizophrenia have mainly focused on positive psychotic symptoms, in particular delusions and hallucinations. The concept of negative symptoms in MA psychosis (e.g., poverty of speech, flatten affect, and loss of drive) is not new but still uncertain. Blunted affect and diminished spontaneity accompanied with paranoid-hallucinatory state were described in 1960s (Yui et al.,
960
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
2000). In a study of 11 MA psychotic inpatients assessed by using the Scale for the Assessment of Negative Symptoms, Japanese investigators found that most patients showed a significant impairment on avolition-apathy, anhedonia-asociality, and attentional impairment (Tomiyama, 1990). Srisurapanont et al. (2003) found negative psychotic symptoms in 21.4% of 168 MA psychotic patients with various ethnicity (Srisurapanont et al., 2003) In this later study, the results of an exploratory factor analysis (EFA) also showed an independent syndrome of negative symptoms with a variance of 26.6%. Despite the above-mentioned evidence, negative symptoms of MA psychosis are still viewed as less severe and/or prevalent (Zorick et al., 2008). Differential item functioning (DIF) statistical techniques are based on the principle that if different groups of patients (e.g., males vs. females) have the same level of disease severity (e.g., depression, psychosis), they should be rated or response similarly on an individual rating item of a measure, regardless of group membership. DIF assessment proceeds by controlling for an estimate of the underlying construct and then examining whether individuals in different groups have a similar distribution of responses to a particular item. The statistical techniques that directly compare item severity between groups are inferior to DIF analyses by the fact that the difference found by a direct comparison may not be a true phenomenological difference. It may be just reflective of greater overall symptom severity in one group versus another. In contrast, the existence of DIF between different groups on a symptom indicates the real difference of symptom severity because the overall disease or syndrome severity has been controlled. Recently, we have finished two studies: (i) the WHO Multi-Site Project on Methamphetamine-Induced Psychosis (WHO-MAIP study), a cross-sectional observation study aiming to evaluate MA psychosis (Ali et al., 2006, Ali et al., 2010, Srisurapanont, et al., 2003), and (ii) Risperidone Long-Acting Injection in Thai Schizophrenic Patients (RLAI-Thai study), a three-month, non-randomized, open-label, single-arm study of risperidone long-acting injection in Thai individuals with schizophrenia (Arunpongpaisal et al., 2010). The psychotic symptoms of participants in both studies were assessed using the Manchester scale (Krawiecka et al., 1977). Because the symptoms of these two studies were comparable, we proposed to apply the DIF statistical techniques to differentiate the severity of psychotic symptoms between these populations. 2. Methods Data of MA psychotic and schizophrenic patients were those of the participants in the WHO-MAIP and the RLAI-Thai studies, respectively. Both studies were carried out in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans. The protocol of each study was approved by the Institution Review Board or the Ethic Committee responsible for the participating hospital or institution. Written informed consent was obtained from each participant after the study details had been fully explained. 2.1. Participants The participants of WHO-MAIP study were MA psychotic patients admitted in many hospitals located in Australia, Japan, the Philippines, and Thailand. In addition to the evidence of MA use during the week prior to the admission, the diagnosis of MA psychosis was confirmed using the Mini-International Neuropsychiatric Interview-Plus (MINIPlus), Module M. The MINI-Plus is a more detailed version of the MINI, a structured clinical interview for the DSM-III-R diagnosis (Sheehan et al., 1998). It includes additional questions to ascertain the relationship between a psychiatric condition and a physical illness
or drug use. These patients were assessed only once within 3–7 days after the admission. The participants of RLAI-Thai study were Thai inpatients and outpatients with DSM-IV schizophrenia, who did not fully respond or could not tolerate their antipsychotic medications. They were willing to switch their previous antipsychotic medications to 25, 37.5, or 50 mg of RLAI given by intramuscular injection. Exclusion criteria included (i) serious unstable medical condition, including recent and present clinically relevant laboratory abnormalities; (ii) history of alcohol dependence or drug abuse (except nicotine) within 3 months of entry into the trial, (iii) on clozapine during the last 3 months; (iv) history or current symptoms of tardive dyskinesia; (v) history of neuroleptic malignant syndrome; (vi) pregnant or breast-feeding female; (vii) female patient of childbearing potential without adequate contraception; (viii) participation in an investigational drug trial in the 30 days prior to selection; and (ix) known intolerance/non-responder to risperidone. Other psychiatric comorbidity was not assessed. Therefore, the patients with comorbidity of other psychiatric illnesses were allowed to participate in this study. Participants were assessed prior to RLAI administration (baseline) and then again after 3 months and 6 months of monthly RLAI treatment. Because the psychotic symptoms at baseline were the most severe ones, the data at this time point were applied in this study. 2.2. Measures In WHO-MAIP study, the participants' psychotic symptoms were assessed using the Manchester scale. This 5-point scale (0 = absent, 1 = mild, 2 = moderate, 3 = marked and 4 = severe) assesses eight psychiatric symptoms commonly found in psychotic patients, including depression, anxiety, delusions, hallucinations, incoherence speech, poverty of speech, flatten/incongruous affect, and psychomotor retardation. For each item of the scale, its inter-rater reliability (Kandall's coefficient of concordance W) is between 0.6 and 0.9 (Krawiecka, et al., 1977). Its concurrent validity has been tested with the Schedules for the Assessment of Positive Symptoms (SAPS) (Andreasen, 1984) and the Schedules for the Assessment of Negative Symptoms (SANS) (Andreasen, 1983). While the Manchester scale items of delusions, hallucinations, and incoherence are significantly correlated with the total SAPS score (r's = 0.53–0.64), its items of poverty of speech, affect flattening, and psychomotor retardation are also significantly correlated with the total SANS score (r's = 0.62–0.85) (Jackson et al., 1990). Except disorganized and catatonic behavior, the Manchester scale covers all DSM-IV schizophrenic symptoms and proposed remission criteria for schizophrenia (American Psychiatric Association, 2000, Andreasen et al., 2005). The modified version of Manchester scale was applied in the RLAIThai study. As used in an antipsychotic trial (Johnstone et al., 1978), the item of flatten/incongruous affect was separated into two items of flatten and incongruous affects. To make the scores of these two items comparable to the single one of flatten/incongruous affect in the WHO-MAIP study, the highest score of these two items in each participant was used as the representative of flatten/incongruous affect item. The different versions of Manchester scale used in both studies were caused by the decision of each investigator team. Both versions were administered by certified psychiatrists, who were well-trained on the use of Manchester scale. 2.3. Data analyses 2.3.1. Unidimensionality The DIF is typically examined on the basis that the set of items is intended to measure a single underlying construct (i.e., the unidimension of items). An examination of the underlying dimension of an item set is therefore the first step in DIF detection. More
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
961
Table 1 Comparisons of demographic data and severity of symptoms between 168 methamphetamine psychotic patients and 169 schizophrenic patients.a Significant differencea
Methamphetamine psychotic patients (n = 168)
Schizophrenic patients (n = 169)
n
n
Sex (male,%)
127 (75.6%) Mean (SD)
114 (67.5%) Mean (SD)
χ2 = 2.74, df = 1, p = 0.12 Mean (SD)
Age Age at first onset of psychotic symptoms Total Manchester scale score Negative syndrome Poverty of speech Psychomotor retardation Flatten/incongruouse affect Positive syndrome Delusions Hallucinations Incoherence speech Anxiety/depression syndrome Depression Anxiety
27.1 25.0 10.7 2.7 0.7 0.8 1.2 5.3 2.1 2.0 1.1 2.8 1.1 1.7
29.8 (14.0) 23.2 (8.4) 8.0 (5.5) 2.9 (2.5) 0.7 (0.9) 0.9 (1.0) 1.3 (1.0) 3.7 (3.0) 1.4 (1.3) 1.4 (1.3) 0.9 (1.0) 1.5 (1.7) 0.5 (0.8) 1.0 (1.1)
t = 2.20, p = 0.03 t = 1.89, p = 0.06 Z = −4.77, p b 0.01 Z = −0.54, p = 0.59 Z = −0.18, p = 0.86 Z = −0.40, p = 0.69 Z = −1.36, p = 0.17 Z = −4.86, p b 0.01 Z = −5.10, p b 0.01 Z = −4.03, p b 0.01 Z = −2.11, p = 0.04 Z = −6.89, p b 0.01 Z = −6.33, p b 0.01 Z = −5.70, p b 0.01
a
(7.6) (9.1) (4.9) (2.3) (0.9) (0.8) (1.0) (2.9) (1.2) (1.4) (1.1) (1.7) (1.0) (1.0)
Except a Chi-square test for sex and a Student's t test for age items, Mann–Whitney U tests were used for all comparisons.
specifically, ones should examine dimensionality separately for each group (i.e., MP or schizophrenic group in the present analyses) by determining if the same number of factors is found in each group (Teresi and Fleishman, 2007). In addition, confirmatory factor analysis (CFA) should be applied to test the single-factor model fit of each item set. The EFA on the eight items of Manchester scale scores obtained from the MA psychotic group was completed in our previous study (Srisurapanont et al., 2003). The same methods were applied on the scores obtained from schizophrenic patients. All eight items were subjected to principal-component analysis for identifying the distinct factors. Eigenvalue one test was used to keep or discard factors. Finally, varimax rotation was performed to elicit the factor components. The EFA was examined by using the SPSS 17.0. After the combination of both data sets, each factor found on the EFA was further confirmed by using the chi-square test of a single-factor model performed by the AMOS 18.0. 2.3.2. DIF analyses We employed the nonparametric kernel-smoothing techniques of item response theory (IRT) implemented in the software (TestGraf) developed by Ramsay (1991, 2001). The strengths of these techniques are that non-parametric IRT models do not require complex
estimation procedures, can be applied to relatively small data sets, are less imposing concerning distributional form of item response functions, and help to avoid misleading results obtained from parametric IRT models. In the present analyses, we determined the item response functions directly from the data without forcing the data to conform to a logistic IRT model. TestGraf analyzes the severity of symptoms observed in the participants of both groups. A response characteristic curve shows how the item score varies as a function of what the test is designed to measure, that is, its latent trait (i.e., severity of each syndrome in the present analyses). To examine DIF, this software calculates a weighted average of the squared difference between the focal group's probability of endorsing an item and the reference group's probability of endorsing an item. Thus, a composite index of DIF is obtained after the comparison of response characteristic curves. For each item, there are two curves of both groups being compared. The curves are numbered in the order in which the files of each group have been selected. The first group is called the focal group (i.e., MA psychotic patients), and the other group is called a reference group (i.e., schizophrenic patients). For each focal group, a summary measure of how far the curve this group is from that of the reference group is calculated. The extent to which the curves differ is called differential item functioning (DIF). If this measure is
Table 2 Exploratory factor analysis of eight-item Manchester scale in 168 methamphetamine psychotic and 169 schizophrenic patients. Methamphetamine psychotic patients (n = 168)a
Schizophrenic patients (n = 169)
Factor 1
Factor 1
Factor 2
Factor 3
Factor 2
Factor 3
0.89 0.88 0.75
-0.17 0.23 0.20
0.10 -0.01 0.39
0.13 0.17 -0.29
-0.01 0.07 0.33
0.56 0.32 -2.16
0.73 0.79 0.77
-0.11 0.27
0.85 0.76
0.10 0.19
0.85 0.85
0.08 0.14
1.92 23.95
1.51 18.81
2.27 28.35
2.00 24.93
1.94 24.24
Negative syndrome Poverty of speech Psychomotor retardation Flatten/incongruous affect
0.86 0.83 0.75
0.07 0.04 0.28
0.02 0.25 0.03
Positive syndrome Delusions Hallucinations Incoherent speech
0.08 0.07 0.32
0.82 0.79 0.67
Anxiety/depression syndrome Depression Anxiety
0.19 0.05 2.13 26.64
Eigenvalue Per cent variance a
(Srisurapanont et al., 2003, reprinted with permission).
962
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
statistically significant, the focal and reference groups are not rated in the same way on that item. In the present analyses, a composite index (measure) of DIF for each focal group was calculated. Taking account the Santor and colleagues' criteria, a value of 0.30 or greater was considered as the existence of DIF, where a value of 0.10 or less suggested little or no DIF (Santor et al., 1994).
A
3. Results 3.1. Characteristics of the participants Of 181 patients participating in the WHO-MAIP study, the Manchester scale scores were completed in 168 participants: 32 participants from Australia, 36 from Japan, 50 from the Philippines, and 50 from Thailand. Their means (SDs) age at first MA use and duration of MA regular use were 19.7 (5.9) years and 3.8 (5.4) years, respectively. The mean (SD) of the maximum amount of MA use during the 12 months prior to the assessment was 26.8 (130.9) g. The data on number of treatment for MA use were available in 140 patients. Of these, 70 patients (50%) had ever received the treatment with a mean (SD) of 1.2 (2.1) times. Of 184 Thai schizophrenic and schizoaffective patients participated in the RLAI-Thai study, nine patients were excluded due to their diagnosis of schizoaffective disorder and the other six participants with schizophrenia were excluded due to the incompletion of Manchester scale scores. Therefore, the data of 169 schizophrenic participants in the RLAI-Thai study were included in the present analyses. Table 1 shows the demographic data and symptom severity of both groups. Of note, MA psychotic group were significantly younger (p = 0.03) but tended to have a later onset of psychotic symptoms (p = 0.06). In addition, their positive and anxiety/depressive syndromes, as well as symptoms, were more severe than those of schizophrenic patients (p's b 0.01 for all comparisons).
B
C
3.2. Factor structure of psychotic symptoms Table 2 shows the previous EFA results of 168 MA psychotic patients and the present results of 169 schizophrenic patients. Although the item of delusions in schizophrenic patients moderately loaded (0.56) on anxiety/depression syndrome, all highest factor loadings of each item were located on the factors as found in the MA psychotic patients. These results suggested a three-factor model of psychotic symptoms in both groups, including a negative syndrome (poverty of speech, psychomotor retardation and flatten/incongruous affect), a positive syndrome (delusions, hallucinations and incoherent speech), and an anxiety/depression syndrome (anxiety and depression). The following chi-square test of single-factor model fit also confirmed the goodness-of-fit of each factor. 3.3. Differential item functioning analysis Figs. 1A–1C, 2A–2C, and 3A and 3B show the response characteristic curves of each item generated from the Manchester scale scores obtained from the MA psychotic and schizophrenic groups. All eight pairs of the curves were almost parallel with all DIF composite indexes lower than 0.3. Except the minimal DIF of incoherent speech (a DIF composite index of 0.142), the composite indexes of the other seven symptoms were lower than 0.1. 4. Discussion 4.1. Main findings The DIF composite indexes of all eight items that were lower than 0.3 indicate the similarity of psychotic symptoms found in MA
Fig. 1. Differential item functioning of negative symptoms: 168 methamphetamine psychotic patients (group 1) vs. 169 schizophrenic patients (group 2) (X axis shows the combined scores of 3 negative symptoms). (Y axis shows the scores of each item). (A) Poverty of speech. (B) Psychomotor retardation. (C) Flatten and incongruous affect.
psychotic and schizophrenic patients. The composite index being lower than 0.1 in the comparisons of seven symptoms suggests that, except the incoherence speech, these seven symptoms, including the negative ones, are rated almost the same in both groups. It is noteworthy that the direct comparison of negative-symptom severity by using a simple statistical technique of Mann–Whitney U test also yielded the same results of nonsignificant differences between groups. Although the direct comparisons of syndrome and symptom severity by using Mann–Whitney U test revealed that MA psychotic
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
A
A
B
B
C
963
Fig. 3. Differential item functioning of anxiety/depression symptoms: 168 methamphetamine psychotic patients (group 1) vs. 169 schizophrenic patients (group 2) (X axis shows the combined scores of 2 anxiety/depression symptoms). (Y axis shows the scores of each item). (A) Depression. (B) Anxiety.
4.2. Relevance of present and previous findings
Fig. 2. Differential item functioning of positive symptoms: 168 methamphetamine psychotic patients (group 1) vs. 169 schizophrenic patients (group 2) (X axis shows the combined scores of 3 positive symptoms). (Y axis shows the scores of each item). (A) Delusions. (B) Hallucinations. (C) Incoherence speech.
patients had more severe positive symptoms (i.e., delusions, hallucinations, incoherence speech), depression, and anxiety (see Table 1), these findings may not be true phenomenological differences. They may be caused by the greater severity of overall positive and anxiety/depression syndromes. This issue is similar to the problems found in many studies directly comparing the severity of bipolar and major depressive symptoms in which Weinstock and colleagues have criticized that the differential symptom expressions reported in those studies may be a reflection of greater overall symptom severity in one group versus another (Weinstock et al., 2010).
The factor structure of MA psychotic and schizophrenic patients found in the present analysis, especially the negative factor, is similar to that of a previous study (Harvey et al., 1996). By using five psychotic symptoms of the Manchester scale, Harvey and colleagues conducted a factor analysis of psychotic symptoms in three groups of schizophrenic patients, including those living in community (n = 320), acute inpatients (n = 49), and long-stay inpatients (n = 35). Their study found that poverty of speech, psychomotor retardation, and flatten/incongruous affect highly loaded in the negative factor, and delusions and hallucination highly loaded in the same factor. To our knowledge, this is the first DIF analysis comparing MA psychotic and schizophrenic symptoms. These findings extend the previous knowledge on prevalence and factor structure of negative symptoms in MA psychotic patients (Srisurapanont et al., 2003; Tomiyama, 1990). The present findings suggest that negative symptoms are not only common in MA psychotic patients but also rated exactly in the same way as those found in schizophrenic patients. In addition, the positive and anxiety/depression symptoms observed in both populations are also similar. 4.3. Limitation of the present analysis There are some limitations of the present study. First, the sample size of 168 or 169 for each group is relatively small for the current concept of DIF analysis. The minimum sample size for logistic regression DIF analysis, which is a statistic technique similar to the
964
M. Srisurapanont et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 959–964
one used in TestGraf, is usually in the range of 100–200 per group (Lai et al., 2005). However, the results of a recent study suggested that a minimum of 200 respondents per group may be needed to ensure the adequate performance (N80% power) (Scott et al., 2009). Second, it is impossible to exclude the effects of antipsychotic medications on the findings of this study. According to the nature of disease, almost all participants in RLAI-Thai study had been taking antipsychotic medications for months or years before the assessment. In the contrary, as patients with short-term psychotic symptoms, many participants in the WHO-MAIP study had been taking antipsychotic medications for only days or weeks. Moreover, a few of them were not taking an antipsychotic medication at all. Third, because the schizophrenic patients in this study were those not responding well or not being able to tolerate some antipsychotic medications, they might not represent the general schizophrenic patients. Fourth, the small number of items included in the Manchester scale may not cover some psychotic symptoms currently described in schizophrenic patients, e.g., disorganized or catatonic behavior. Last, the ethnicities of participants in both groups are not the same. While the MA psychotic groups composed of Australians, Japanese, the Filipinos and Thai, only Thai patients participated in the RLAI-Thai study. Whether the ethnicity difference had an impact on the expression of psychotic symptoms remains unknown. 4.4. Implications of the findings Despite the above-mentioned limitations, the findings that psychotic symptoms in MA psychotic and schizophrenic patients are almost the same may suggest a share of pathophysiology of MA psychosis and schizophrenia. The present study's results, therefore, support a recent proposal of using amphetamine psychosis as a human model for studying the onset and course of schizophrenia (Hermens et al., 2009). Although this human model is still underutilized, the concept of using amphetamine as an environmental pathogen may be another strategy for the study of schizophrenia. 4.5. Conclusion The psychotic symptoms, including the negative ones, found in MA psychotic and schizophrenic patients are almost the same. More specifically, at the same level of syndrome severity (i.e., negative, positive, and anxiety/depression syndromes), the severity of each symptom observed in both groups are not much different. Further studies in large sample sizes are warranted. Conflict of interest None. Acknowledgments We wish to thank coauthors of the WHO-MAIP and the RLAI-Thai studies who allowed us to reanalyze the data.
References Ali R, Baigent M, Marsden J, Montiero M, Srisurapanont M, Sunga A, et al. WHO MultiSite Project on Methamphetamine-induced Psychosis: A descriptive report of findings from participating countries. Adelaide (South Australia): Drug & Alcohol Services South Australia; 2006. Ali R, Marsden J, Srisurapanont M, Sunga A, Baigent M, Monteiro M. Methamphetamine psychosis in Australia, Philippines, and Thailand: Recommendations for acute care and clinical inpatient management. Addict Disord Their Treat 2010;9:143–9. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4-TR ed. Washington (DC): American Psychiatric Association; 2000. Andreasen NC. The scale for the assessment of negative symptoms (SANS). Iowa City: University of Iowa; 1983. Andreasen NC. The scale for the assessment of positive symptoms (SAPS). Iowa City: University of Iowa; 1984. Andreasen NC, Carpenter JWT, Kane JM, Lasser RA, Marder SR, Weinberger DR. Remission in schizophrenia: proposed criteria and rationale for consensus. Am J Psychiatry 2005;162:441–9. Arunpongpaisal S, Srisurapanont M, Kongsakon R, Kitiwattanagul K, Samanwongthai U. Risperidone long-acting injection (RLAI): the 12-week efficacy and tolerability in Thai patients with chronic schizophrenia. J Med Assoc Thai 2010;93:343–50. Bell DS. Comparison of amphetamine psychosis and schizophrenia. Br J Psychiatry 1965;111:701–7. Griffith JD, Cavanaugh J, Held J, Oates J. Dextroamphetamine: evaluation of psychomimetic properties in man. Arch Gen Psychiatry 1972;26:97-100. Harvey C, Curson D, Pantelis C, Taylor J, Barnes T. Four behavioural syndromes of schizophrenia. Br J Psychiatry 1996;168:562–70. Hermens DF, Lubman DI, Ward PB, Naismith SL, Hickie IB. Amphetamine psychosis: a model for studying the onset and course of psychosis. MJA 2009;190:S22–5. Jackson HJ, Burgess PM, Minas IH, Joshua SD. Psychometric properties of the Manchester scale. Acta Psychiatr Scand 1990;81:108–13. Johnstone EC, Crow TJ, Frith CD, Carney MW, Price JS. Mechanism of the antipsychotic effect in the treatment of acute schizophrenia. Lancet 1978;1(8069):848–51. Krawiecka M, Goldberg D, Vaughan M. A standardized psychiatric assessment scale for rating chronic psychotic patients. Acta Psychiatr Scand 1977;55:299–308. Lai JS, Teresi J, Gershon R. Procedures for the analysis of differential item functioning (DIF) for small sample size. Eval Health Prof 2005;28:283–94. Ramsay JO. Kernel smoothing approaches to non parametric item curve estimation. Psychometrika 1991;56:611–30. Ramsay JO. TesGraf: A program for the graphical analysis of multiple choice test and questionnaire data. Montreal, Canada: McGill University; 2001. Robinson T, Becker JB. Enduring changes in brain and behavior produced by chronic amphetamine administration: a review and evaluation of animal models of amphetamine psychosis. Brain Res 1986;396:157–98. Santor DA, Ramsay JO, Suroff DC. Nonparametric item analyses of the Beck Depression Inventory: evaluating gender item bias and response option weights. Psychol Assess 1994;6:255–70. Scott NW, Fayers PM, Aaronson NK, Bottomley A, de Graeff A, Groenvold M, et al. A simulation study provided sample size guidance for differential item functioning (DIF) studies using short scales. J Clin Epidemiol 2009;62:288–95. Sheehan D, Lecrubier Y, Sheehan K, Amorim P, Janavs J, Weiller E, et al. The MiniInternational Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59:22–33. Srisurapanont M, Ali R, Marsden J, Sunga A, Wada K, Montiero M. Psychotic symptoms in methamphetamine psychotic in-patients. Int J Neuropsychopharmacol 2003;6:347–52. Teresi JA, Fleishman JA. Differential item functioing and health assessment. Qual Life Res 2007;16:33–42. Tomiyama G. Chronic schizophrenia-like states in methamphetamine psychosis. Jpn J Psychiatry Neurol 1990;44:531–9. Weinstock LM, Strong D, Uebelacker LA, Miller IW. DSM-IV depressive symptom expression among individuals with a history of hypomania: a comparison to those with or without a history of mania. J Psychiatr Res 2010;44:979–85. Yui K, Ikemoto S, Ishiguro T, Goto K. Studies of amphetamine or methamphetamine psychosis in Japan. Ann NY Acad Sci 2000;914:1-12. Zorick TS, Rad D, Rim C, Tsuang J. An overview of methamphetamine-induced psychotic syndromes. Addict Disord Their Treat 2008;7:143–56.