Clozapine Underutilization and Discontinuation in African Americans
Due to Leucopenia
Deanna L. Kelly1, Julie Kreyenbuhl, Lisa Dixon,
TheMarylandPsychiatric Research Center, Box21247,Baltimore,
MD 21228 and the Center for Mental Health Services Research
Clozapine use has been notably lower in African American
patients than in Caucasians. It has been suggested that
lower normal ranges for white blood cell (WBC) counts
in African Americans, known as benign ethnic neutropenia,
may account partially for the disparity. We examined the
rates of leucopenia and agranulocytosis as reasons for dis-
continuation of clozapine in a sample of 1875 patients with
schizophrenia treated in the State of Maryland. Between
1989 and 1999, 5.3% (31/588) of African Americans and
2.4% (31/1287) of Caucasians discontinued clozapine
treatment due to leucopenia (chi square = 10.35, df = 1,
ulocytosis while 8 Caucasian patients (0.62%) developed
this blood dyscrasia. Discontinuations due to leucopenia
occurred throughout treatment. Discontinuations due to
agranulocytosis occurred primarily in the first 18 weeks
(7/8; 87.5% patients with agranulocytosis). It is likely
that African Americans had clozapine discontinued unnec-
essarily due to benign ethnic neutropenia. We concur with
recent recommendations to acknowledge differences in
WBC values in African Americans and to modify prescrib-
ing guidelines or formally acknowledge benign ethnic leu-
copenia like in other countries in order to facilitate greater
use of clozapine in these patients.
Key words: clozapine/white blood cells/eucopenia/
agranulocytosis/benign ethnic neutropenia/race
Approximately 1.5 million people in the United States
suffer from schizophrenia, and between 20–30% of these
patients are unresponsive to first-line antipsychotic ther-
apy. Clozapine is the only antipsychotic medication
approved for treatment-resistant schizophrenia. The su-
periority of this second-line antipsychotic compared with
phrenia is well established.1Additionally, recent findings
from the clinical antipsychotic trials of intervention
effectiveness II trial demonstrate the superiority of cloza-
pine over other second-generation antipsychotics2in
a population of patients who prospectively failed an op-
timized antipsychotic trial.
Given the high costs of medication discontinuation,
rehospitalization and inadequate treatments for schizo-
phrenia, the underutilization of clozapine in the United
States in particular is noteworthy.3,4It is likely that clo-
zapine’s underutilization is related in part to its frequent
monitoring and serious side effects including agranulocy-
tosis, myocarditis, other inflammatory reactions, seiz-
ures, obesity, diabetes mellitus, and other metabolic
abnormalities. Moreover, racial disparities in the use
of clozapine have been consistently observed, with Afri-
can Americans less likely to receive this medication than
Caucasians.5–7Possible explanations for this prescribing
pattern include prescriber bias, the anticipation of non-
adherence, and the notion of lower effectiveness in Afri-
can Americans. The superior effectiveness of clozapine
relative to other antipsychotics merits increased efforts
to encourage greater use in appropriate patients and to
more efficiently monitor for side effects.2Earlier work
by our group had also found lower rates of clozapine pre-
scribing among inpatients with schizophrenia who are
African American as compared with Caucasians. We
racial disparity and found that differences in presenting
symptoms and response potential do not explain the ap-
parent underutilization of clozapine in African Ameri-
cans relative to Caucasians.7
A hypothesized explanation for the disparity in cloza-
pine prescribing may be related to the requirements for
white blood cell (WBC) monitoring. Current guidelines
require WBCs to be >3500/mm3before clozapine is initi-
ated, and the drug must be discontinued if WBCs fall be-
low 3000/mm3or if the absolute neutrophil count (ANC)
goes below 1500/mm3(red alert zone). However, normal
WBC levels in African American males generally range
from 2800 to 9500/mm3, which is significantly lower
1To whom correspondence should be addressed; tel: 717-764-
9260, fax: 410-402-6038, e-mail: email@example.com.
Schizophrenia Bulletin vol. 33 no. 5 pp. 1221–1224, 2007
Advance Access publication on December 14, 2006
? The Author 2006. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved.
For permissions, please email: firstname.lastname@example.org.
than the normal range for Caucasians (3600–9500/mm3)
and also overlaps with the red alert zone, a range of
WBC that requires stopping clozapine.8Due to this phe-
nomenon, known as benign ethnic neutropenia, it has
icans may be deemed ineligible for clozapine. Further, up
to one quarter of African American patients may have
clozapine discontinued for this reason.9,10In fact, in
in African Americans have been challenged.9,11Thus, the
aim of the current study was to compare the rates of clo-
zapine discontinuation and reasons for discontinuation
including leucopenia or agranulocytosis in a large popu-
lation of African American and Caucasian patients.
thorization and Monitoring Program (CAMP) database
from the State of Maryland. Between January 1, 1989,
and December 31, 1999, 2136 patients began clozapine
therapy. Clozapine patients had a Diagnostic and Statis-
tical ManualofMental Disorders, FourthEdition, diagno-
at least 2 different chemical classes of antipsychotic med-
ing clinician, and had a total score of $35 or a score of
$4 on selected individual items of the Brief Psychiatric
Rating Scale. Clozapine use was contraindicated in
an uncontrolled seizure disorder, a WBC <3500/mm3,
a history of myeloproliferative disorder, or pregnancy.
All patients contained in the CAMP data were those
who were treated with medical or pharmacy assistance.
Of the 2136 patients, 1875 patients with race data
(African American or Caucasian) were included in this
study. A small sample of Hispanic, Asian, and Pacific
Islanders were excluded and not included in the analysis.
Reasons for clozapine discontinuation were categorized
as lack of efficacy, nonadherence, agranulocytosis, leuco-
penia, other hematologic adverse effect, other nonhema-
tologic adverse effect, death, and unknown. All sites
utilizing clozapine were required to report reasons for
discontinuation and a clinical pharmacist followed up
to ensure consistent reporting. According to manufac-
tinuation due to the WBC profile, leucopenia was a WBC
<3000/mm3or ANC of <1500/mm3. Agranulocytosis
was defined as an ANC <500/mm3. Patients who discon-
tinued clozapine for <7 days were not considered as dis-
continued from clozapine. This study was approved by
the State of Maryland and the University of Maryland
Institutional Reviews Boards, and a waiver of informed
consent was granted due to the nature of the data.
We examined differences by race in rates of clozapine
discontinuation using chi squares. Student t tests were
used for analyzing differences in continuous variables.
All analyses were 2-tailed, and significance was defined
as alpha < 0.05.
ican patients initiated treatment with clozapine. Overall,
864/1875 (46.1%) discontinued clozapine at some point
during their first clozapine trial. Similar proportions of
African American (48.6%) and Caucasian patients
(44.9%) discontinued clozapine for any reason. As shown
in table 1, discontinuation for leucopenia and for un-
known reasons were significantly higher in African
Americans compared with Caucasians.
A significantly higher percentage of males in the Afri-
can American group were noted to discontinue clozapine
due to leucopenia as compared with the Caucasian co-
hort. (80.6% vs 45.2%; chi square = 8.36, df = 1, P =
0.0038). No differences in age were noted (38.1 6 12.9
and 40.1 6 12.1 years; t = 0.63, df = 60, P = 0.53). The
percent of African Americans and Caucasians who dis-
continued in the first 18 weeks was 58.1% (18/31) and
38.7% (12/31), respectively. Between 18 and 52 weeks
29.0% (9/31) of African Americans and 32.3% (10/31)
of Caucasians discontinued. After >52 weeks, 29.0%
(9/31) and 32.3% (8/31) of the groups discontinued cloza-
pine (chi square = 3.18, df = 2, P = 0.20).
No African American patients developed agranulocy-
tosis while 8 Caucasian patients (0.62%) developed this
blood dyscrasia. The total rate of agranulocytosis for
both races was 8/1875 (0.43%). Of the 8 Caucasians
who discontinued due to agranulocytosis, 5/8 (62.5%)
were males, and the mean age was 48.1 6 10.1 years.
Seven of 8 (87.5%) Caucasians who developed agranulo-
cytosis did so in the first 18 weeks of treatment (mean:
68.3 6 30.2 days).
Our study presents the first report to our knowledge de-
scribing rates of leucopenia and agranulocytosis among
persons receiving clozapine by race in a large cohort. We
show that significantly more African American patients
casians (2.4%). In fact, this was the only reason that dif-
unknown reasons. It is possible that the some of the un-
documented reasons (unknown), which were higher in
African Americans, may have also been leucopenia cases.
It appeared that African American men may be at the
greatest risk for discontinuation, and this is supported
by other literature describing lower normal WBC counts
occurring in African American men compared with
women.8Also of interest was that no cases of agranulo-
D. L. Kelly et al.
higher vigilance and higher discontinuations by clinicians
occurrence of agranulocytosis which is possibly geneti-
patients were discontinued unnecessarily due to the be-
nign ethnic neutropenia present in this racial group.
Previous work by our group and others has described
disparities in clozapine treatment in African Americans
relative to Caucasians.7While the reasons for this contin-
ued disparity remain unclear, it may be mediated in part
by guidelines for WBC monitoring. Values greater than
3500/mm3are required before clozapine treatment can be
initiated. Yet, many African Americans exhibit benign
ethnic neutropenia, in which the lower ranges of their
WBC fall within the red alert zone.8–12This could poten-
tially lead to underprescribing of clozapine or its unnec-
essary discontinuation in African American patients.
It is important to note that low baseline WBCs have
not been associated with agranulocytosis among cloza-
pine patients, and agranulocytosis does not occur more
frequently in African Americans.10In fact, evidence sug-
gests that patients continuing treatment in the red alert
zone appear to maintain their WBCs or have their
WBCs recover in most instances following a drop below
the threshold.13Patients with morning pseudoneutrope-
nia have also been maintained successfully on clozapine
with no risk for agranulocytosis.14As such, recent pub-
lications have challenged the prescribing guidelines in
minority patients. In fact, some have argued in favor of
creating different treatment recommendations for minor-
ity groups now that a body of data has been established
detailing the risk of agranulocytosis.9,11Additionally,
Novartis has recently revised their prescribing guidelines
in the United Kingdom and Canada to include special
consideration for patients with benign ethnic neutrope-
nia.11Despite these observations, acknowledgments,
and changes in labeling in other countries, a change to
current recommendations for African Americans in the
United States has not yet occurred.
As we have observed, the risk for leucopenia can occur
at any point during treatment. On the other hand, agran-
of cases developing the blood dyscrasia during this time-
3rd month of treatment.15It is unknown if the African
Americans who developed leucopenia would have pro-
gressed to agranulocytosis. Our study is limited by the
cases (<3000/mm3) be discontinued from clozapine, and
this reason was documented in the electronic database.
This study suggests the need for further research on the
appropriateness of the clozapine WBC guidelines for
African Americans. Recent labeling changes in other
countries underscore the possibility that modification of
prescribing recommendations for clozapine in the United
use of clozapine in African American populations.
This project was supported by pilot funding from the
Health Disparities Research Grant (NIH P60, Wilson D,
Principal Investigator (PI)) and an Advanced Center
for Intervention and Services Research (P50 MH40279,
Carpenter WT Jr, PI). The authors would like to thank
Yang Yu, MS for data analysis and the members of the
Mental Health Disparities Steering Committee.
1. Conley RC, Kelly DL. Management of treatment-resistant
schizophrenia. Biol Psychiatry. 2001;50:898–911.
Table 1. Reasons for Discontinuation by Racial Group
African American (%),
N = 588
N = 1287Test Statistic
Total48.6, N = 28644.9, N = 578 Chi square = 2.26, df = 1, P = 0.13
Lack of efficacy 9.18, N = 5411.27, N = 145 Chi square = 1.85, df = 1, P = 0.17
Agranulocytosis0 0.62, N = 8Chi square = 3.67, df = 1, P = 0.06
Other hematologic adverse effecta
5.27, N = 31 2.41, N = 31Chi square = 10.35, df = 1, P = 0.001
0.34, N = 20.31, N = 4Chi square = 0.01, df = 1, P = 0.92
Other nonhematologic adverse effect5.27, N = 316.22, N = 80 Chi square = 0.65, df = 1, P = 0.42
Nonadherence14.6, N = 8612.5, N = 161Chi square = 1.58, df = 1, P = 0.21
Death1.36, N = 81.63, N = 21 Chi square = 0.19, df = 1, P = 0.66
Other3.23, N = 193.41, N = 44Chi square = 0.04, df = 1, P = 0.83
Unknown9.35, N = 556.52, N = 84Chi square = 4.70, df = 1, P = 0.03
athrombocytopenia, low complete blood count, eosinophilia.
Leucopenia, Race, and Clozapine Discontinuation
antipsychotic treatment. Am J Psychiatry. 2006;163:600–610.
3. Fayak M, Flowers C, Signorelli D, Simpson G. Underuse of
evidence-based treatments in psychiatry. Psychiatr Serv.
4. Conley RR, Kelly DL, Lambert TJ, Love RC. A comparison
of clozapine use in Maryland and in Victoria, Australia. Psy-
chiatr Serv. 2005;56:320–323.
5. Mallinger JB, Fisher SG, Brown T, Lamberti JS. Racial dispa-
rities in the use of second-generation antipsychotics for the
treatment of schizophrenia. Psychiatr Serv. 2006;57:133–136.
6. Copeland LA, Zeber JE, Valenstein M, Blow FC. Racial dis-
parity in the use of atypical antipsychotic medications among
veterans. Am J Psychiatry. 2003;160:1817–1822.
7. Kelly DL, Dixon LB, Kreyenbuhl JA, et al. Clozapine utiliza-
tion and outcomes by race in a public mental health system
1994–2000. J Clin Psychiatry. 2006;67:1404–1411.
8. Bain BJ. Ethnic and sex differences in the total and differen-
tial white cell count and platelet count. J Clin Pathol. 1996;
9. Fisher N, Baigent B. Treatment with clozapine. Black
patients’ low white cell counts currently mean that they can-
not be treated. BMJ. 1996;313:1262.
10. Munroe J, O’Sullivan D, Andrews C, Arana A, Mortimer A,
Kerwin R. Active monitoring of 12,760 clozapine recipients in
the UK and Ireland. Beyond pharmacovigilance. Br J Psychi-
11. Mallinger JB, Lamberti JS. Clozapine—should race affect
prescribing guidelines? Schizophr Res. 2006;83:107–108.
12. Rajagopal S. Clozapine, agranulocytosis, and benign ethnic
neutropenia. Postgrad Med. 2005;81:545–546.
13. Ahn YM, Jeong SH, Jang HS, et al. Experience of maintain-
ing clozapine medication in patients with ‘‘red-alert zone’’
neutropenia: long-term follow-up results. Int Clin Psycho-
14. Esposito D, Chouinard G, Hardy P, Corruble E. Success-
ful initiation of clozapine treatment despite morning
pseudoneutropenia. Int J Neuropsychopharmacol. 2006;9:
JA. Clozapine-induced agranulocytosis: incidence and risk
factors in the United States. N Engl J Med. 1993;329:162–167.
D. L. Kelly et al.