ORIGINAL RESEARCH
Cocaine Use in Individuals With Schizophrenia Impact on Doses of Discharge Antipsychotic Medications Satyajit Mohite, MBBS, Ikenna Ngana, MBBS, and Olaoluwa O. Okusaga, MD, MScPHR
Objectives: Despite the high prevalence of cocaine use disorder in schizophrenia, the impact of cocaine on antipsychotic requirement has not been studied in this population. The aim of this study was to evaluate the effect of cocaine on doses of antipsychotic medication prescribed during periods of acute exacerbation of psychotic symptoms in individuals with schizophrenia. Methods: We reviewed the medical records of individuals with schizophrenia discharged from hospitals between 2008 and 2012. Student t tests and linear regression were used to compare doses of discharge antipsychotic medications (in chlorpromazine equivalents) between individuals with schizophrenia with cocaine positive urine drug test results (n = 180; age 42.71 ± 10.03 years) and individuals with schizophrenia with negative urine drug test results (n = 3194; age 38.49 ± 12.86 years). Results: Unadjusted analysis revealed that individuals with schizophrenia who tested positive for cocaine were discharged on lower doses of antipsychotic medication compared with those who tested negative (449.88 ± 2.12 vs 515.47 ± 2.16; P = 0.021). However, after adjusting for age, sex, race, and length of stay, the 2 groups did not differ on doses of discharge antipsychotic medication (geometric mean difference 7.41; CI: 7.62-12.30; P = 0.703). Conclusions: Our preliminary result suggests that cocaine use does not impact significantly on the doses of antipsychotic medication prescribed during periods of acute exacerbation of psychosis in schizophrenia and individuals with schizophrenia with comorbid cocaine use disorder may require similar doses of antipsychotic medication as those without cocaine use disorder. Key Words: antipsychotic medication doses, cocaine, cocaine use disorder, chlorpromazine equivalents, schizophrenia (J Addict Med 2015;9: 177–180)
From the University of Texas Harris County Psychiatric Center (SM, IN, OOO), School of Public Health (SM), and Department of Psychiatry and Behavioral Sciences (OOO), The University of Texas Health Science Center, Houston. Received for publication June 14, 2014; accepted December 22, 2014. The authors declare no conflicts of interest. Send correspondence and reprint requests to Olaoluwa O. Okusaga, MD, MScPHR, University of Texas–Harris County Psychiatric Center, 2800 South MacGregor Way, Houston, TX 77021. E-mail: [email protected]. C 2015 American Society of Addiction Medicine Copyright ISSN: 1932-0620/15/0903-0177 DOI: 10.1097/ADM.0000000000000110
S
chizophrenia, a severe mental illness with an estimated worldwide prevalence of 4/1000 (Saha et al., 2005), is also highly comorbid with cocaine use disorder (CUD) and 12% to 44% of individuals with schizophrenia have been reported to meet criteria for CUD (Sevy et al., 1990; Dixon et al., 1991; Khalsa et al., 1991; Galanter et al., 1992; Elangovan et al., 1993; Shaner et al., 1993; Mueser et al., 2000). The preference for cocaine in individuals with schizophrenia is reflected in a recent outpatient study in which the authors reported that in individuals with dual diagnoses of substance use disorder and psychotic disorder, “50% used cocaine, 21.4% opioids, 12.5% alcohol, 10.7% cannabis, 3.6% designer drugs, and 1.8% benzodiazepines” (Grau-L´opez et al., 2014, page 86). Because dopamine is the central neurotransmitter implicated in cocaine’s reinforcing properties and in the pathogenesis of symptoms of schizophrenia, cocaine use can be expected to impact significantly on symptoms of schizophrenia (Dickson et al., 1995). Cocaine-using individuals with schizophrenia have been found to have more impairment in verbal memory relative to non–cocaine-using patients (Sevy et al., 1990). Exacerbation of psychotic symptoms, more hostility, worsened depression, higher rates of hospitalization, less prosocial behavior, poorer cognition, higher medication nonadherence, poorer insight, homelessness, violent offending, and increased risk of death have all been associated with cocaine abuse in schizophrenia (Sevy et al., 1990; Dickson et al., 1995; Schmidt et al., 2011; Addy et al., 2012). Cocaine-using individuals with schizophrenia also seem to have low motivation for reduction of substance use, along with lower psychological functioning in the context of work, school, or homemaking. Remission rates of CUD are lower in this patient population when compared with other substances, and this makes treatment more challenging in this population (Addy et al., 2012). The specific underlying mechanisms of the association of CUD with schizophrenia are unclear. Some of the hypotheses put forward in an attempt to explain this association include self-medication, differential emotional valence, reward dysfunction, and genetic makeup (Addy et al., 2012). With regard to the treatment and management of patients with a dual diagnosis of schizophrenia and CUD, no well-established pharmacological intervention currently exists. Although evidence exists to suggest that atypical antipsychotics may have some utility in the treatment of comorbid CUD in schizophrenia (Sabioni et al., 2013), there is a dearth of studies designed to compare doses of antipsychotics in individuals with
J Addict Med r Volume 9, Number 3, May/June 2015
Copyright © 2015 American Society of Addiction Medicine. Unauthorized reproduction of this article is prohibited.
177
J Addict Med r Volume 9, Number 3, May/June 2015
Mohite et al.
schizophrenia with CUD and those without CUD. We have therefore aimed to compare doses of antipsychotic medication at the time of discharge from hospital, between individuals with schizophrenia with urine drug screen (UDS) test results positive for cocaine (CP) and those with negative UDS results (CN). On the basis of our expectation that individuals with schizophrenia with CUD will be more difficult to treat during periods of exacerbation of psychosis, we hypothesized that CP individuals with schizophrenia will be discharged on higher doses of antipsychotic medication when compared with CN individuals with schizophrenia.
METHODS Population and Setting We retrieved information from medical records of adult patients (n = 5106) discharged from hospital with a diagnosis of schizophrenia between January 2008 and December 2012. Data gathered included information on age, sex, race, antipsychotic medications, length of stay in hospital, and diagnoses. The diagnosis of schizophrenia was made by the psychiatrist responsible for the care of the patient during the admission. The diagnosis of schizophrenia was based on the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSMIV) (American Psychiatric Association, 2000) criteria. To reduce the likelihood of including cases of substance-induced psychosis, we excluded individuals with a discharge diagnosis of schizophreniform disorder or psychotic disorder not otherwise specified. All the patients were treated with antipsychotic medications, the doses of which were converted to chlorpromazine equivalents (Andreasen et al., 2010). Haloperidol and fluphenazine were the 2 commonly prescribed typical antipsychotic medications in this sample. Olanzapine, risperidone, quetiapine, ziprasidone, aripiprazole, and clozapine were the atypical antipsychotics prescribed. Paliperidone, iloperidone, asenapine, and lurasidone were not on formulary in the hospital when this study was conducted. We also gathered data on the results of urine drug tests administered on the patients at the time of admission. Urine drug screens are routine on admission and this study did not change care or level of risk. The institutional review board of the University of Texas Health Science Center at Houston approved the study.
Urine Drug Testing Immunoassay was used for the initial screening of the urine specimens obtained at the time of admission to hospital to determine the presence or absence of cocaine. Gas chromatography/mass spectrometry was used to confirm the presence of cocaine and to quantify the amount (Goldberger and Cone, 1994). The screening cutoff concentration for cocaine via immunoassay was 300 ng/mL and 150 ng/mL for the (gas chromatography/mass spectrometry) confirmatory analysis (Kim et al., 2013).
Statistical Analysis Chlorpromazine equivalent and hospital length of stay did not have a normal distribution (they were both skewed to the right) and we therefore applied logarithmic transformation in an attempt to normalize these variables. We compared demographic and clinical variables between CP and CN indi-
178
viduals with schizophrenia, using t tests and chi-square tests. We also used t tests to compare chlorpromazine equivalent doses between CP and CN individuals with schizophrenia. Using linear regression, we compared chlorpromazine equivalent doses between the 2 patient groups, while adjusting for differences in age, sex, race, and length of stay between the 2 groups. Geometric means with 95% confidence intervals are presented for the difference in chlorpromazine equivalents between the 2 groups. The geometric means were obtained by calculating the antilog of the log-transformed mean differences. The analyses were repeated after excluding readmissions. All significance levels reported are 2-sided with P < 0.05, considered statistically significant. Statistical analyses were carried out using IBM SPSS version 20 (IBM CorP, Armonk, NY).
RESULTS Demographic and Clinical Characteristics of the Sample From January 2008 to December 2012, 5106 individuals with schizophrenia were discharged from hospital; of the 5106, 4138 patients had complete data. After we had excluded patients with UDSs positive for other street drugs besides cocaine, a total of 3374 patients (180 positive for cocaine and 3194 negative for any street drug) were remaining and these were the ones included in the final analyses. There was no difference in the sex distribution between the CP and CN groups (P = 0.795), but the CP group was on average older than the CN group (42.71 ± 10.03 years vs 38.49 ± 12.86 years; P < 0.0001) (Table 1). There were more black patients in the CP group (82.2% vs 57.7%; χ 2 = 53.17; P < 0.0001) and more white (21.3% vs 10.0%; χ 2 = 53.17; P < 0.0001) and Hispanic (15.9% vs 6.7%; χ 2 = 53.17; P < 0.0001) patients in the CN group. Although Asians were more in the CN group (4.7% vs 1.1%; χ 2 = 53.17; P < 0.0001), Asians and other populations were underrepresented in both groups. The geometric mean length of stay in hospital was significantly longer in the CN group (11.83 ± 1.83 days vs 8.07 ± 1.92 days; P < 0.0001).
Comparison of Discharge Medication Doses Between the 2 Patient Groups Unadjusted analysis revealed that the CP group was discharged on lower doses of antipsychotic medication compared with the CN group (geometric mean chlorpromazine equivalent doses 449.88 ± 2.12 vs 515.47 ± 2.16; P = 0.021). However, after adjusting for age, sex, racial differences, and length of stay, the difference in geometric mean chlorpromazine equivalent doses between the 2 patient groups was no longer significant (geometric mean difference 7.41; CI: 7.62−12.30; P = 0.703). Similar results were obtained after excluding readmissions.
DISCUSSION Contrary to our hypothesis, we did not observe any difference in doses of antipsychotic medication (at the time of discharge from hospital) between CP and CN individuals with schizophrenia. The decision to focus on discharge medication doses was based on the fact that oftentimes, the dose of antipsychotic medication an individual with schizophrenia receives when leaving the hospital is likely to be the maximum dose C
2015 American Society of Addiction Medicine
Copyright © 2015 American Society of Addiction Medicine. Unauthorized reproduction of this article is prohibited.
J Addict Med r Volume 9, Number 3, May/June 2015
Cocaine Use in Individuals With Schizophrenia
TABLE 1. Characteristics of Schizophrenia Patients With Cocaine-Positive and Cocaine-Negative Urine Drug Screens at Hospital Admission
Demographic and Clinical Variables Age, y Sex, n (%) Male Female Race, n (%)† White Black Hispanic Asian Others Geometric mean length of stay, d Geometric mean chlorpromazine equivalent dose
Schizophrenia Patients With Cocaine Positive UDS (n = 180)
Schizophrenia Patients With Cocaine Negative UDS (n = 3194)
P*
42.71 ± 10.03
38.49 ± 12.86
Cocaine Use in Individuals With Schizophrenia Impact on Doses of Discharge Antipsychotic Medications Satyajit Mohite, MBBS, Ikenna Ngana, MBBS, and Olaoluwa O. Okusaga, MD, MScPHR
Objectives: Despite the high prevalence of cocaine use disorder in schizophrenia, the impact of cocaine on antipsychotic requirement has not been studied in this population. The aim of this study was to evaluate the effect of cocaine on doses of antipsychotic medication prescribed during periods of acute exacerbation of psychotic symptoms in individuals with schizophrenia. Methods: We reviewed the medical records of individuals with schizophrenia discharged from hospitals between 2008 and 2012. Student t tests and linear regression were used to compare doses of discharge antipsychotic medications (in chlorpromazine equivalents) between individuals with schizophrenia with cocaine positive urine drug test results (n = 180; age 42.71 ± 10.03 years) and individuals with schizophrenia with negative urine drug test results (n = 3194; age 38.49 ± 12.86 years). Results: Unadjusted analysis revealed that individuals with schizophrenia who tested positive for cocaine were discharged on lower doses of antipsychotic medication compared with those who tested negative (449.88 ± 2.12 vs 515.47 ± 2.16; P = 0.021). However, after adjusting for age, sex, race, and length of stay, the 2 groups did not differ on doses of discharge antipsychotic medication (geometric mean difference 7.41; CI: 7.62-12.30; P = 0.703). Conclusions: Our preliminary result suggests that cocaine use does not impact significantly on the doses of antipsychotic medication prescribed during periods of acute exacerbation of psychosis in schizophrenia and individuals with schizophrenia with comorbid cocaine use disorder may require similar doses of antipsychotic medication as those without cocaine use disorder. Key Words: antipsychotic medication doses, cocaine, cocaine use disorder, chlorpromazine equivalents, schizophrenia (J Addict Med 2015;9: 177–180)
From the University of Texas Harris County Psychiatric Center (SM, IN, OOO), School of Public Health (SM), and Department of Psychiatry and Behavioral Sciences (OOO), The University of Texas Health Science Center, Houston. Received for publication June 14, 2014; accepted December 22, 2014. The authors declare no conflicts of interest. Send correspondence and reprint requests to Olaoluwa O. Okusaga, MD, MScPHR, University of Texas–Harris County Psychiatric Center, 2800 South MacGregor Way, Houston, TX 77021. E-mail: [email protected]. C 2015 American Society of Addiction Medicine Copyright ISSN: 1932-0620/15/0903-0177 DOI: 10.1097/ADM.0000000000000110
S
chizophrenia, a severe mental illness with an estimated worldwide prevalence of 4/1000 (Saha et al., 2005), is also highly comorbid with cocaine use disorder (CUD) and 12% to 44% of individuals with schizophrenia have been reported to meet criteria for CUD (Sevy et al., 1990; Dixon et al., 1991; Khalsa et al., 1991; Galanter et al., 1992; Elangovan et al., 1993; Shaner et al., 1993; Mueser et al., 2000). The preference for cocaine in individuals with schizophrenia is reflected in a recent outpatient study in which the authors reported that in individuals with dual diagnoses of substance use disorder and psychotic disorder, “50% used cocaine, 21.4% opioids, 12.5% alcohol, 10.7% cannabis, 3.6% designer drugs, and 1.8% benzodiazepines” (Grau-L´opez et al., 2014, page 86). Because dopamine is the central neurotransmitter implicated in cocaine’s reinforcing properties and in the pathogenesis of symptoms of schizophrenia, cocaine use can be expected to impact significantly on symptoms of schizophrenia (Dickson et al., 1995). Cocaine-using individuals with schizophrenia have been found to have more impairment in verbal memory relative to non–cocaine-using patients (Sevy et al., 1990). Exacerbation of psychotic symptoms, more hostility, worsened depression, higher rates of hospitalization, less prosocial behavior, poorer cognition, higher medication nonadherence, poorer insight, homelessness, violent offending, and increased risk of death have all been associated with cocaine abuse in schizophrenia (Sevy et al., 1990; Dickson et al., 1995; Schmidt et al., 2011; Addy et al., 2012). Cocaine-using individuals with schizophrenia also seem to have low motivation for reduction of substance use, along with lower psychological functioning in the context of work, school, or homemaking. Remission rates of CUD are lower in this patient population when compared with other substances, and this makes treatment more challenging in this population (Addy et al., 2012). The specific underlying mechanisms of the association of CUD with schizophrenia are unclear. Some of the hypotheses put forward in an attempt to explain this association include self-medication, differential emotional valence, reward dysfunction, and genetic makeup (Addy et al., 2012). With regard to the treatment and management of patients with a dual diagnosis of schizophrenia and CUD, no well-established pharmacological intervention currently exists. Although evidence exists to suggest that atypical antipsychotics may have some utility in the treatment of comorbid CUD in schizophrenia (Sabioni et al., 2013), there is a dearth of studies designed to compare doses of antipsychotics in individuals with
J Addict Med r Volume 9, Number 3, May/June 2015
Copyright © 2015 American Society of Addiction Medicine. Unauthorized reproduction of this article is prohibited.
177
J Addict Med r Volume 9, Number 3, May/June 2015
Mohite et al.
schizophrenia with CUD and those without CUD. We have therefore aimed to compare doses of antipsychotic medication at the time of discharge from hospital, between individuals with schizophrenia with urine drug screen (UDS) test results positive for cocaine (CP) and those with negative UDS results (CN). On the basis of our expectation that individuals with schizophrenia with CUD will be more difficult to treat during periods of exacerbation of psychosis, we hypothesized that CP individuals with schizophrenia will be discharged on higher doses of antipsychotic medication when compared with CN individuals with schizophrenia.
METHODS Population and Setting We retrieved information from medical records of adult patients (n = 5106) discharged from hospital with a diagnosis of schizophrenia between January 2008 and December 2012. Data gathered included information on age, sex, race, antipsychotic medications, length of stay in hospital, and diagnoses. The diagnosis of schizophrenia was made by the psychiatrist responsible for the care of the patient during the admission. The diagnosis of schizophrenia was based on the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSMIV) (American Psychiatric Association, 2000) criteria. To reduce the likelihood of including cases of substance-induced psychosis, we excluded individuals with a discharge diagnosis of schizophreniform disorder or psychotic disorder not otherwise specified. All the patients were treated with antipsychotic medications, the doses of which were converted to chlorpromazine equivalents (Andreasen et al., 2010). Haloperidol and fluphenazine were the 2 commonly prescribed typical antipsychotic medications in this sample. Olanzapine, risperidone, quetiapine, ziprasidone, aripiprazole, and clozapine were the atypical antipsychotics prescribed. Paliperidone, iloperidone, asenapine, and lurasidone were not on formulary in the hospital when this study was conducted. We also gathered data on the results of urine drug tests administered on the patients at the time of admission. Urine drug screens are routine on admission and this study did not change care or level of risk. The institutional review board of the University of Texas Health Science Center at Houston approved the study.
Urine Drug Testing Immunoassay was used for the initial screening of the urine specimens obtained at the time of admission to hospital to determine the presence or absence of cocaine. Gas chromatography/mass spectrometry was used to confirm the presence of cocaine and to quantify the amount (Goldberger and Cone, 1994). The screening cutoff concentration for cocaine via immunoassay was 300 ng/mL and 150 ng/mL for the (gas chromatography/mass spectrometry) confirmatory analysis (Kim et al., 2013).
Statistical Analysis Chlorpromazine equivalent and hospital length of stay did not have a normal distribution (they were both skewed to the right) and we therefore applied logarithmic transformation in an attempt to normalize these variables. We compared demographic and clinical variables between CP and CN indi-
178
viduals with schizophrenia, using t tests and chi-square tests. We also used t tests to compare chlorpromazine equivalent doses between CP and CN individuals with schizophrenia. Using linear regression, we compared chlorpromazine equivalent doses between the 2 patient groups, while adjusting for differences in age, sex, race, and length of stay between the 2 groups. Geometric means with 95% confidence intervals are presented for the difference in chlorpromazine equivalents between the 2 groups. The geometric means were obtained by calculating the antilog of the log-transformed mean differences. The analyses were repeated after excluding readmissions. All significance levels reported are 2-sided with P < 0.05, considered statistically significant. Statistical analyses were carried out using IBM SPSS version 20 (IBM CorP, Armonk, NY).
RESULTS Demographic and Clinical Characteristics of the Sample From January 2008 to December 2012, 5106 individuals with schizophrenia were discharged from hospital; of the 5106, 4138 patients had complete data. After we had excluded patients with UDSs positive for other street drugs besides cocaine, a total of 3374 patients (180 positive for cocaine and 3194 negative for any street drug) were remaining and these were the ones included in the final analyses. There was no difference in the sex distribution between the CP and CN groups (P = 0.795), but the CP group was on average older than the CN group (42.71 ± 10.03 years vs 38.49 ± 12.86 years; P < 0.0001) (Table 1). There were more black patients in the CP group (82.2% vs 57.7%; χ 2 = 53.17; P < 0.0001) and more white (21.3% vs 10.0%; χ 2 = 53.17; P < 0.0001) and Hispanic (15.9% vs 6.7%; χ 2 = 53.17; P < 0.0001) patients in the CN group. Although Asians were more in the CN group (4.7% vs 1.1%; χ 2 = 53.17; P < 0.0001), Asians and other populations were underrepresented in both groups. The geometric mean length of stay in hospital was significantly longer in the CN group (11.83 ± 1.83 days vs 8.07 ± 1.92 days; P < 0.0001).
Comparison of Discharge Medication Doses Between the 2 Patient Groups Unadjusted analysis revealed that the CP group was discharged on lower doses of antipsychotic medication compared with the CN group (geometric mean chlorpromazine equivalent doses 449.88 ± 2.12 vs 515.47 ± 2.16; P = 0.021). However, after adjusting for age, sex, racial differences, and length of stay, the difference in geometric mean chlorpromazine equivalent doses between the 2 patient groups was no longer significant (geometric mean difference 7.41; CI: 7.62−12.30; P = 0.703). Similar results were obtained after excluding readmissions.
DISCUSSION Contrary to our hypothesis, we did not observe any difference in doses of antipsychotic medication (at the time of discharge from hospital) between CP and CN individuals with schizophrenia. The decision to focus on discharge medication doses was based on the fact that oftentimes, the dose of antipsychotic medication an individual with schizophrenia receives when leaving the hospital is likely to be the maximum dose C
2015 American Society of Addiction Medicine
Copyright © 2015 American Society of Addiction Medicine. Unauthorized reproduction of this article is prohibited.
J Addict Med r Volume 9, Number 3, May/June 2015
Cocaine Use in Individuals With Schizophrenia
TABLE 1. Characteristics of Schizophrenia Patients With Cocaine-Positive and Cocaine-Negative Urine Drug Screens at Hospital Admission
Demographic and Clinical Variables Age, y Sex, n (%) Male Female Race, n (%)† White Black Hispanic Asian Others Geometric mean length of stay, d Geometric mean chlorpromazine equivalent dose
Schizophrenia Patients With Cocaine Positive UDS (n = 180)
Schizophrenia Patients With Cocaine Negative UDS (n = 3194)
P*
42.71 ± 10.03
38.49 ± 12.86
