BRIEF REPORTS

BRIEF
AmREPORTS
J Psychiatry 154:7, July 1997

Olfactory Identification Deficits in Schizophrenia:

Correlation With Duration of Illness

Paul J. Moberg, Ph.D., Richard L. Doty, Ph.D., Bruce I. Turetsky, M.D.,

Steven E. Arnold, M.D., Richard N. Mahr, Psy.D., Ruben C. Gur, Ph.D.,
Warren Bilker, Ph.D., and Raquel E. Gur, M.D., Ph.D.

Objective: The authors examined the relationship between deficits in olfactory identification
and duration of illness in young and elderly patients with schizophrenia. Method: Olfactory
identification performance of 38 patients with schizophrenia and 40 normal subjects was
compared by using the University of Pennsylvania Smell Identification Test. Results: The
schizophrenic patients demonstrated olfactory deficits relative to the comparison group, and
the elderly schizophrenic patients displayed a greater magnitude of olfactory deficit than the
younger patients. Independent of normal aging effects and cognitive deficit, patients with
schizophrenia showed a strong relationship between olfactory identification scores and dura-
tion of illness, which suggests that olfactory abilities decline progressively over the course of
the disorder. Conclusions: In contrast to other neuropsychological measures that have been
reported to be stable over the course of illness, olfactory identification abilities deteriorate
steadily in patients with schizophrenia, even for those with relatively recent onset.
(Am J Psychiatry 1997; 154:1016–1018)

C ross-sectional and longitudinal studies that have

investigated age-related differences in cognitive
performance in patients with schizophrenia have noted
either no differences or a very slow and gradual pro-
gression (1, 2). Deficits in neuropsychological function-
ing have been interpreted as being consistent with a
neurodevelopmental process rather than a neurodegen-
erative model of illness. Deficits in odor identification,
odor detection threshold sensitivity, and odor memory
have been described in young patients with schizophre-
nia (3, 4). Neuroleptic use, smoking, cognitive deficits,
and illness severity all appear unrelated to this abnor-
mality (5). Olfactory processing is mediated by limbic
neuroanatomical structures that have been implicated
in the pathophysiology of schizophrenia, particularly
the prefrontal cortex, ventromedial temporal lobe, ba-
sal forebrain, and diencephalon (6). Olfactory deficits
in patients with schizophrenia may reflect disturbances
of cortical or subcortical brain regions untapped by tra-
Received Feb. 13, 1996; revisions received Sept. 20, 1996, and Jan.
13, 1997; accepted Jan. 17, 1997. From the Departments of Psychia-
try, Otorhinolaryngology, and Biostatistics and Epidemiology, Uni-
versity of Pennsylvania School of Medicine. Address reprint requests
to Dr. Moberg, Brain-Behavior Laboratory, Department of Psychia-
try, Hospital of the University of Pennsylvania, 10th Floor, Gates
Building, 3400 Spruce St., Philadelphia, PA 19104.
Supported by NIMH grant MH-34880, National Institute on Deaf-
ness and Other Communication Disorders grant PO1-00161, and the
EJLB Foundation.
ditional neuropsychological measures. No study has yet
examined the effect of aging or duration of illness on
olfactory function in patients with schizophrenia.
METHOD
Thirty-eight patients who met DSM-III-R criteria for schizophre-
nia and 40 healthy volunteers were recruited by the Mental Health
Clinical Research Center on Schizophrenia at the University of Penn-
sylvania. All subjects underwent a medical and psychiatric evaluation
that included a physical examination and the administration of stand-
ardized clinical scales. Exclusion criteria included history of psychiat-
ric disorder (other than schizophrenia), neurologic disorder, head
trauma, loss of consciousness, substance abuse, or the presence of one
of the following: a medical condition that could alter cerebral func-
tioning, an upper respiratory infection, or a condition that could af-
fect olfactory functioning (e.g., common cold). Normal subjects were
screened for psychiatric history in first-degree relatives. Written in-
formed consent was obtained from all subjects.
Of the younger patients with schizophrenia (age range=18–50
years, N=20), 10 were neuroleptic-naive, and 10 were taking neuro-
leptics. All elderly patients (age range=66–88, N=18) were taking
neuroleptic medication. The schizophrenic patients did not differ
from the comparison subjects in age (schizophrenic group: mean=
50.6 years, SD=25.5; comparison group: mean=49.6 years, SD=24.6),
gender composition (schizophrenic group: 18 men and 20 women;
comparison group: 18 men and 22 women), race (84% [N=32] of the
schizophrenic patients and 73% [N=29] of the comparison subjects
were Caucasian), or smoking history (schizophrenic group: 18 non-
smokers, 13 past smokers, and seven current smokers; comparison
group: 23 nonsmokers, 11 past smokers, and six current smokers).
Olfactory identification was assessed with the University of
Pennsylvania Smell Identification Test (7), a forced-choice test with

1016 Am J Psychiatry 154:7, July 1997


four response alternatives for each of 40 odors. The Picture Identi-
fication Test was also administered. This test is identical to the
smell identification test except that line drawings are presented in-
stead of odors to screen for cognitive deficits that may confound
results (8). The schizophrenic patients and normal subjects did not
differ in performance on the Picture Identification Test (F=0.62,
df=1, 76, p=0.43). General cognitive status was assessed with the
Mini-Mental State Examination (9).
RESULTS
Scores on the University of Pennsylvania Smell Identi-
fication Test for the schizophrenic patients differed from
those of the comparison group and indicated significant
olfactory identification deficits (F=37.3, df=1, 74,
p<0.001). There was no observed main effect for gender
or diagnosis-by-gender interaction (all p values >0.40). In
the younger schizophrenic patients, scores of those who
were neuroleptic-naive did not differ from scores of those
who were medicated, which suggests minimal medication
effects (p=0.88). Analysis of covariance between younger
and elderly schizophrenic patients for performance on the
University of Pennsylvania Smell Identification Test, with
age and gender as the covariates, revealed that the elderly
patients demonstrated greater olfactory deficit than the
younger patients (F=13.5, df=1, 34, p<0.001).
A significant correlation was seen between scores on
the University of Pennsylvania Smell Identification Test
and duration of illness, which was defined as the length
of time since the onset of psychotic symptoms in con-
text of functional decline (r=–0.88, df=36, p<0.001).
Poorer scores were associated with longer duration of
schizophrenia. To ensure that this relationship was not
confounded by normal age and gender effects, we ad-
justed the University of Pennsylvania Smell Identifica-
tion Test values to account for these influences by using
regression coefficients that were computed from the
normal comparison group. Results indicated that even
when variance attributable to normal age and gender
effects was removed, the relationship with duration of
illness remained (r=–0.92, df=36, p<0.001) (figure 1).
To assess the impact of general cognitive deficits on
performance, partial correlations were calculated be-
tween scores on the University of Pennsylvania Smell
Identification Test and duration of illness with Mini-Men-
tal State scores held constant. Analysis of residuals re-
vealed that controlling the effect of generalized cognitive
impairment did not diminish the correlation between
scores on the University of Pennsylvania Smell Identifica-
tion Test and duration of illness (r=–0.92, df=34, p<
0.001). Similarly, patients did not score below impair-
ment cutoffs on the Picture Identification Test, which in-
dicated that they had the cognitive skills to visually iden-
tify the test items but could not identify the odor.
DISCUSSION
This is the first study to demonstrate that elderly pa-
tients with schizophrenia have an olfactory identifica-
tion deficit relative to age-matched comparison subjects
Am J Psychiatry 154:7, July 1997
BRIEF REPORTS
FIGURE 1. Relation of Age- and Gender-Adjusted Scores on the Uni-
versity of Pennsylvania Smell Identification Test to Duration of Ill-
ness in 38 Patients With Schizophrenia
and that the magnitude of this deficit is greater than
that seen in younger schizophrenic patients. While lon-
gitudinal assessment of individual patients is the ideal
method to demonstrate decline over the lifespan, this is
rarely possible for human studies. However, these
cross-sectional data strongly suggest that in patients
with schizophrenia, olfactory identification perform-
ance decreases over the course of the illness until the
very late stages of life. Indeed, there seemed to be a cou-
pling between performance on the University of Penn-
sylvania Smell Identification Test and duration of ill-
ness that was independent of normal aging and gender
effects. The tightness of this coupling permits predic-
tion of illness duration from olfactory performance and
raises the possibility that olfactory structures are al-
tered by slow neurodegenerative processes in schizo-
phrenia. Two caveats must be noted. First, the elderly
patients had been taking neuroleptic medication for
many years. While effects of long-term neuroleptic
treatment on olfaction are not fully understood, pub-
lished empirical data that concerned the effects of medi-
cation on University of Pennsylvania Smell Identifica-
tion Test performance in several neurological disorders
have been uniformly negative. For schizophrenia spe-
cifically no study has noted differences among neuro-
leptic-naive, neuroleptic-withdrawn, and currently
medicated patients on olfactory test performance (4, 5).
Second, fewer elderly male than female subjects were
available for study, which perhaps militates against fur-
ther elucidation of sex differences in test performance.
Recent neuropathological studies of younger and eld-
erly schizophrenic patients have described a variety of
cytoarchitectural abnormalities in brain regions associ-
ated with olfactory functioning (6, 10). A postmortem
study that examined astrocytosis (pathology thought to
result from postmaturational CNS injury) in elderly
schizophrenic patients found modest increases in astro-
1017
BRIEF REPORTS
cytosis in hippocampal and orbitofrontal cortices, re-
gions known to be involved in olfaction (11). To our
knowledge, no specific neuropathological investigation
of primary or unimodal olfactory structures in schizo-
phrenia has been conducted. We hypothesize, however,
that it would reveal pathological changes in olfactory
brain regions. On the basis of these data, such investi-
gation is clearly warranted.
REFERENCES
1. Heaton R, Paulson JS, McAdams LA, Kuck J, Zisook S, Braff D,
Harris MJ, Jeste DV: Neuropsychological deficits in schizo-
phrenics: relationship to age, chronicity, and dementia. Arch
Gen Psychiatry 1994; 51:469–476
2. Sweeney JA, Haas GL, Keilp JG, Long M: Evaluation of the sta-
bility of neuropsychological functioning after acute episodes of
schizophrenia: one-year follow-up study. Psychiatry Res 1991;
38:63–76
3. Geddes J, Huws R, Pratt P: Olfactory acuity in the positive and
negative syndromes of schizophrenia. Biol Psychiatry 1991; 29:
774–778
4. Wu J, Buchsbaum MS, Moy K, Denlea N, Kesslak P, Tseng H,
1018
Plosnaj D, Hetu M, Potkin S, Bracha S, Cotman C: Olfactory
memory in unmedicated schizophrenics. Schizophr Res 1993; 9:
41–47
5. Kopala LC, Clark C, Hurwitz T: Olfactory deficits in neuroleptic
naive patients with schizophrenia. Schizophr Res 1992; 8:245–
250
6. Arnold SE, Trojanowski JQ: Recent advances in defining the neu-
ropathology of schizophrenia. Acta Neuropathol (Berl) 1996; 92:
217–231
7. Doty RL, Shaman P, Dann M: Development of the University of
Pennsylvania Smell Identification Test: a standardized microen-
capsulated test of olfactory function. Physiol Behav 1984; 32:
489–502
8. Vollmecke T, Doty RL: Development of the Picture Identifica-
tion Test (PIT): a research companion to the University of Penn-
sylvania Smell Identification Test. Chem Senses 1985; 10:413–
414
9. Folstein MF, Folstein SE, McHugh PR: “Mini-Mental State”: a
practical method for grading the cognitive state of patients for
the clinician. J Psychiatr Res 1975; 12:189–198
10. Bogerts B: Recent advances in the neuropathology of schizophre-
nia. Schizophr Bull 1993; 19:431–435
11. Arnold SE, Franz BR, Trojanowski JQ, Moberg PJ, Gur RE: Gli-
al fibrillary acidic protein-immunoreactive astrocytosis in elderly
patients with schizophrenia and dementia. Acta Neuropathol
1996; 91:269–277
Am J Psychiatry 154:7, July 1997