This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Morrow, L. A. Articles by Scott, A. Search for Related Content PubMed PubMed Citation Articles by Morrow, L. A. Articles by Scott, A. Related Collections Anxiety Depression Stress and Coping

Increased Incidence of Anxiety and Depressive Disorders in Persons With Organic Solvent Exposure

From the Department of Psychiatry (L.A.M., G.R.B., L.S., R.C.), University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; Department of Psychology (C.G.), Long Island University, Brooklyn, New York; and Armstrong Occupational Health Care (A.S.), Armstrong County Memorial Hospital, Kittanning, Pennsylvania.

Address reprint requests to: Lisa Morrow, PhD, Western Psychiatric Institute and Clinic, 3811 O’Hara St., Pittsburgh, PA 15213. Email: morrowla{at}msx.upmc.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
OBJECTIVE: The purpose of this study was to determine whether the prevalence of current and past DSM-IV axis I psychiatric disorders is higher among persons with a history of exposure to organic solvents than among a demographically similar group of nonexposed control subjects.

METHODS: Thirty-eight solvent-exposed subjects and 39 nonexposed healthy control subjects were evaluated for axis I disorder with the Structured Clinical Interview for DSM-IV.

RESULTS: A significantly higher number of solvent-exposed subjects (71%) met criteria for current DSM-IV axis I disorder in comparison with control subjects (10%). The most prevalent diagnosis in exposed subjects was within the anxiety and mood clusters, with a high percentage (36%) of exposed subjects meeting criteria for a dual diagnosis of mood and anxiety disorder. There were no differences between the groups in past psychiatric disorders or current or past substance abuse or dependence.

CONCLUSIONS: The rates of past psychiatric disorders among solvent-exposed subjects are similar to those among normal control subjects, but the prevalence of current DSM-IV axis I psychiatric disorders is significantly higher among exposed subjects than among control subjects.

Key Words: organic solvents, • depression • anxiety, • psychiatric diagnoses • DSM-IV.

Abbreviations: DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, fourth edition; SCID = Structured Clinical Interview for DSM-IV.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
For more than a century, reports in the scientific literature have documented alterations in psychiatric symptomatology in persons with a history of exposure to organic solvents (1, 2). Cross-sectional epidemiological studies, using ratings, interviews, personality inventories, and behavior checklists, have found higher frequencies of psychiatric complaints among solvent-exposed workers compared with nonexposed workers (3–9). The presence of such symptomatology has been shown for both workers currently exposed to solvents but with no physical or mental complaints (ie, field studies) as well as persons presenting to a medical clinic with complaints of adverse health effects related to chronic long-term exposure or an acute overexposure (ie, clinical studies) (4, 6, 10). Research to date has not, however, used structured clinical interviews to assess diagnoses in accordance with DSM-IV (11), nor has previous research assessed the history of psychiatric disorders in this population.

The purpose of the present study was three-fold: 1) to compare the rate of current DSM-IV psychiatric disorder in persons with exposure to organic solvents with the rate in nonexposed control subjects, 2) to assess the existence of past axis I psychiatric diagnoses, and 3) to examine whether certain diagnostic classes are more prominent in exposed individuals. Accordingly, we predicted, on the basis of past research (5, 6), that exposed persons would have higher rates of current diagnoses and that the diagnoses would be clustered within the mood disorders and anxiety disorders. Finally, we hypothesized that there would be no difference in past psychiatric diagnoses between the exposed and nonexposed groups.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
The current study was part of a larger study conducted at the University of Pittsburgh and VA Healthcare System and funded by the National Institute of Mental Health. The purpose of the larger study was to address cognitive, psychiatric, and neurophysiological sequelae in workers with a history of exposure to organic solvents. The present analyses focused on the psychiatric assessment of current and past axis I disorders.

Participants
All solvent-exposed subjects who participated in this study were referred from the Department of Environmental and Occupational Health, University of Pittsburgh School of Medicine, over a 31/2-year span (July 1995 through December 1998). The large majority of patients seen in the Environmental and Occupational Health Clinic had been exposed in the workplace and were generally referred to the clinic by their primary care physician. Control subjects were recruited through newspaper advertisement and flyers posted in the community. The advertisements stated recruitment was for healthy persons "between the ages of 30 and 55 who do not have a history of exposure to organic solvents." Forty solvent-exposed subjects and 44 nonexposed control subjects provided informed consent. Exposed subjects were excluded if they reported current symptoms or treatment regimens that would interfere with the reliability of the cognitive and neurophysiological testing procedures used in the larger research protocol, and nonexposed subjects were excluded if on subsequent interviews they reported current or past solvent exposure. Specifically, two solvent-exposed subjects and five control subjects were excluded for the following reasons: history of treatment for a past episode of schizophrenia and current psychotic symptoms (one male exposed subject), current medication regimen that interfered with attention and memory processes (one male exposed subject), current psychotic symptoms (two male control subjects), past occupational exposure to solvents (one male control subject), presumed drug use at the time of testing (one male control subject), and borderline level of intellectual functioning (ie, IQ < 80; one female control subject).

The remaining 38 solvent-exposed and 39 nonexposed subjects completed a comprehensive assessment, including history of chemical exposure and psychiatric diagnostic evaluation. The exposed group consisted of 33 men and 5 women with a mean age of 43.8 years (SD = 10.29) and 12.2 years of education (SD = 2.2). The control group consisted of 23 men and 16 women without a history of chemical exposure. Control subjects had a mean age of 40.1 years (SD = 11.0) and 13.4 years of education (SD = 1.7). Control subjects received identical evaluations. The groups did not differ significantly in age (p = .203). Education was higher in the control group (p = .05). We also calculated socioeconomic status using the Hollingshead Index (12). There were no differences between groups in this index (mean values: exposed group = 32.8, control group = 33.1; NS). All subjects were evaluated only after giving informed consent. Subjects were paid a nominal fee for participation in the protocol. The institutional review boards of the University of Pittsburgh School of Medicine and the VA Healthcare System approved the study.

Within the exposed group, the mean duration of exposure was 8.26 years (range = 1 day to 30 years). All subjects reported exposure to mixtures of organic solvents in the workplace and at the time of the assessment indicated residual physical and mental changes (eg, headaches, fatigue, and poor memory). The most common exposure was to the aromatic and halogenated hydrocarbons (eg, toluene, xylene, benzene, and trichloroethylene). Additionally, some subjects reported secondary exposure to other chemicals (eg, hydrogen sulfide). The primary route of exposure was inhalation, although many had direct dermal contact with the solvents. The average time between exposure and assessment was 22 months (range = 1 week to 9 years). Sixteen percent of the subjects reported they had an episode of peak exposure, that is, exposure to an excessive amount of solvent that resulted in seeking treatment from a doctor or hospitalization. Given that there was a wide range of exposure histories, subjects were placed into one of three groups based on the severity and duration of exposure by the first author (L.A.M.). Those placed in group 1 had the mildest exposures (ie, shortest duration of exposure and no peak exposure). Those in group 3 had the longest duration of exposure and/or a history of peak exposure resulting in treatment from a physician or hospital. Persons who fell in between were placed in group 2.

Measures
The psychiatric diagnostic evaluation consisted of the SCID for DSM-IV (13). Current and past (ie, lifetime) axis I disorders were determined in the standard manner. For example, queries to determine the presence or absence of past major depressive disorder were worded according to a standardized script (ie, "Have you ever had a period when you were feeling depressed or down most of the day, nearly every day?"). Psychiatric interviews were conducted face to face in the typical manner by trained research staff; the interviews were supervised by one of the authors (R.C.). Interviewers were not blind to exposure status. Clinical interview training was consistent with previous studies of clinical interviewing (14). The interviewer scored symptoms as usual, and then scores were entered using the computer-assisted clinical version of SCID (15). This computer-scoring package was developed by the original authors of the SCID and includes more than 37 specific axis I disorders and has the same psychometric properties as the clinical interview version.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Table 1 presents the frequencies for current and past axis I psychiatric disorders for the exposed and nonexposed groups. For the purposes of statistical analysis, we included the diagnoses of generalized anxiety, obsessive-compulsive disorder, and phobia under the broad category of anxiety. Likewise, mood disorder consisted of the general category that combined major depressive disorder, depressive disorder not otherwise specified, and bipolar disorder. This approach was taken because alterations in mood and anxiety are the most common finding for solvent-exposed persons and there is no a priori reason to assume that this population would exhibit a specific subtype of mood or anxiety disorder. Moreover, the relatively small numbers within the subgroups would have precluded meaningful interpretation of the findings.

There were no differences between the two groups in past psychiatric disorders. Rates of past mood disorder, anxiety, and substance abuse or dependence were similar for the two groups; that is, six solvent-exposed subjects and six control subjects met criteria for past mood disorder (² = 0.078, p = .519). Likewise, there were also no significant differences in rates of past anxiety (² = 0.320, p = .510) or substance abuse or dependence between the two groups (² = 0.005, p = .571).

Within the exposed group, five subjects had a past axis I diagnosis that preceded their solvent exposure. Of these five persons, only three met criteria for current axis I diagnosis, and the other two had no current diagnoses at the time of assessment. The above findings remained the same after reanalysis of the data in which these four subjects were excluded. That is, there were significant differences between exposed and nonexposed subjects for current axis I diagnoses (73 vs. 10%) as well as specific diagnoses in the mood (52 vs. 2.5%), anxiety (57 vs. 7%), and panic (27 vs. 0%) categories (all p values < .001). No differences in current substance abuse (5.8 vs. 2.5%) were found, nor were there any differences between the groups for past axis I diagnoses.

Because there was a higher number of female control subjects than female exposed subjects, we looked at the rate of past psychiatric diagnoses between men and women in the control group. There were no differences in rates of men and women in the control group meeting criteria for a past psychiatric diagnosis: Seven women and eight men met criteria. Furthermore, the diagnoses were spread across domains for men and women and did not center within the mood, anxiety, or substance abuse or dependence clusters.

Within the exposed group, ratings for exposure intensity (1–3) were assessed in relation to diagnoses. Although the number of persons in each rating group was small (group 1, N = 14; group 2, N = 19; group 3, N = 5), there was evidence of a higher proportion of diagnoses related to group. That is, for a diagnosis of anxiety, 43% of persons in group 1 met criteria, whereas 63 and 80% of subjects in groups 2 and 3, respectively, met criteria. Similar trends were noted for mood disorder (43, 52, and 60%, respectively) and panic (14, 32, and 40%, respectively). Tests for the difference between proportions, however, did not reach significance (all p values > .05).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
These results are in agreement with past research documenting alterations in psychiatric symptomatology in persons with a history of exposure to organic solvents (3–9). No previous studies have, however, used structured clinical interviews to assess DSM-IV criteria to categorically classify psychiatric disorder in this population. As predicted, there was a significantly higher rate of current DSM-IV axis I diagnoses in the exposed group compared with the control group. This finding remained even after excluding five exposed subjects who had a past axis I diagnosis that preceded their exposure. In the current sample, 71% of exposed persons met criteria for a current axis I diagnosis, whereas only 10% of control subjects met criteria. Within the exposed sample, the most prevalent diagnosis was noted within the clusters of anxiety and mood: 58% of this sample met criteria for a diagnosis of anxiety, and 50% met criteria for a diagnosis of mood disorder. However, a large number of persons in the exposed group (36%) had both disorders. The high percentage of comorbidity is not surprising, because comorbid axis I disorders are a common phenomenon in depressed individuals, especially a dual diagnosis that includes anxiety disorder (16). There were, however, no differences between the groups on rates of current substance abuse or dependence. This latter finding is somewhat unexpected in light of the usual association of increased risk for comorbid disorders of substance abuse and psychiatric disorder in the general population (17).

Perhaps the most striking finding from the present study is the similarity of the two groups with regard to diagnoses for past psychiatric symptomatology, particularly with respect to mood and anxiety disorders. Solvent-exposed subjects did not suffer from higher rates of past psychiatric disturbances compared with nonexposed control subjects; for the most part, the rates of past disorder were small and almost identical for the two groups. In contrast to the rates for current diagnoses, where there was an increased proportion of anxiety disorder, only one subject in the exposed group and one in the nonexposed group met criteria for past anxiety disorder. Moreover, a criterion for past mood disorder was noted in only three exposed subjects and three control subjects, although three additional persons in each group were diagnosed with comorbid disorders of past mood disorder and substance abuse or dependence or phobia. Thus, the present findings provide no evidence that solvent-exposed patients have an increased incidence of past psychiatric disorder that could account for the heightened rates of psychiatric symptomatology that has frequently been reported for this population.

Because of the variation in exposure duration within the exposed group, we classified persons into three groups based on their exposure history. Those with the mildest exposure were placed in 1, those with the most severe exposure were placed in group 3, and those in between were placed in group 2. Although not significant, the results showed that group 3 had the highest percentage of persons meeting criteria for the three primary diagnoses (anxiety, depression, and panic). This is in general agreement with field studies, using self-report indices, which have found an association between solvent exposure and increased psychiatric symptomatology (10) and recent symptoms of depression (4).

The pathogenesis of psychiatric dysfunction in solvent-exposed patients has typically been attributed to either a traumatic psychological reaction to the exposure (18) or secondary to neurological damage with resultant mood disorder (19). Consistent with a possible underlying neurological disruption, a number of recent studies have found evidence of alterations of functional brain metabolism in the frontotemporal areas in persons with a history of significant solvent exposure (20, 28). This is also consistent with research implicating frontal and frontotemporal areas in the pathogenesis of affective disorders such as depression (21, 22) and anxiety (23). From the current study we cannot conclude with any degree of certainty that the mood and anxiety disorders are attributable to underlying neurological disruption or determine whether the disorders are solvent induced. However, the findings suggest that future studies should investigate the possibility that alteration of functional brain abnormalities due to past exposure may play a role in the psychiatric disorder seen in solvent-exposed patients.

There are several caveats with regard to the current data. First, our control subjects had a somewhat higher level of education than exposed subjects. However, there were no differences between the groups on SES as measured by the Hollingshead Index (12). Thus, differences in the rates of current axis I disorders are unlikely to be due to disparities between the groups in basic sociodemographic factors. Second, we used a control group consisting of healthy nonexposed adults from the surrounding community with rates of axis I diagnoses somewhat similar, although not identical, to those published for US residents (24). That is, our control subjects had rates of "affective" and "addictive" disorders analogous to those in the National Comorbidity Survey: 10 vs. 14% for affective disorders and 31 vs. 35% for addictive disorders, respectively. Our control sample, however, had lower rates of "anxiety" disorders (2.6 vs. 22%). The latter finding may be due to the relatively small size of our sample. Also, interviewers were not blind to exposure status. One way to resolve this issue in future studies is to have two sets of interviewers. One interviewer would query for relevant information concerning psychiatric symptoms in relation to exposure and occupation, and a second interviewer, blinded to the subject’s occupational history, would query for psychiatric symptoms not related to occupational history. In addition, because of the way the SCID is administered and differences in queries for different disorders, onset of current axis I diagnosis in relationship to onset of exposure was not systematically determined. It has been suggested that one of the first indicators of solvent encephalopathy is change in psychiatric status (10). Our data provide evidence of a high prevalence of ongoing psychiatric disorder in persons presenting to an occupational medicine clinic with past workplace exposure. However, our current data do not allow us to determine whether there is a specific time after exposure is initiated that these changes begin to occur. Modification of the SCID to include a time line of symptoms in relation to exposure is recommended for future studies. Finally, because the exposed patients were referred from an occupational medicine clinic, generalizations to other exposed groups (eg, exposed workers with no symptoms) are limited. It would be of interest in future studies to compare rates of diagnoses in clinical and nonclinical samples of exposed workers and to compare these with rates in similar nonexposed community samples.

From these results, we can conclude that this patient population is at risk for a high prevalence of psychiatric problems. Because there are more than 9 million workers with exposure to solvents in the workplace (25), the potential for the development of substantial psychiatric disorders secondary to such exposure is great. If such individuals are not effectively treated, their psychiatric symptoms may prove extremely costly, both in healthcare costs and in absenteeism and decreased workplace productivity (26). Consequently, there is a definite need for continued research in this area. Future studies may consider using various control groups, such as a medically ill blue-collar sample without exposure. Specific research that addresses intervention strategies for this population is also needed. There has been some positive evidence supporting the use of specialized cognitive remediation strategies in treating solvent-exposed patients (27). However, to our knowledge, no studies have looked at pharmacologic intervention with this population, which would seem especially warranted given the findings of this study.

	ACKNOWLEDGMENTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
This research was supported by Grant MH-43846 from the National Institutes of Mental Health. The authors thank Amy Lederer and Dan Massaro for assistance with various aspects of data collection.

Received for publication October 7, 1999.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 

1. Hartman DE. Neuropsychological toxicology: identification and assessment of human neurotoxic syndromes. 2nd ed. New York: Plenum Press; 1995.
2. Morrow LA, Muldoon SB, Sandstrom DJ. Neuropsychological sequelae associated with occupational and environmental exposure to chemicals. In: Tarter R, Butters MA, Beers SR, editors. Medical neuropsychology: the impact of disease on behavior. 2nd ed. New York: Plenum Press; in press.
3. Baker E, Letz R, Eisen E, Pothierl EA, Plantamura DL, Larson M, Wolford R. Neurobehavioral effects of solvents in construction painters. J Occup Med 1988; 30: 116–23.[Medline]
4. Bolla KI, Schwartz BS, Agnew J, Ford PD, Bleecker ML. Subclinical neuropsychiatric effects of chronic low-level solvent exposure in US paint manufacturers. J Occup Med 1990; 32: 671–7.[Medline]
5. Morrow LA, Ryan CM, Goldstein G, Hodgson MJ. A distinct pattern of personality disturbance following exposure to organic solvents. J Occup Med 1989; 31: 743–6.[Medline]
6. Morrow LA, Kamis H, Hodgson M. Psychiatric symptomatology in persons with organic solvent exposure. J Consult Clin Psychol 1993; 61: 171–4.[Medline]
7. Struwe G, Wennberg A. Psychiatric and neurological symptoms in workers occupationally exposed to organic solvents: results of a differential epidemiological study. Acta Psychiatr Scand 1983; 67: 68–80.
8. Struwe G, Mindus P, Johnson B. Psychiatric ratings in occupational health research: a study of mental symptoms in lacquerers. Am J Ind Med 1980; 1: 23–30.[Medline]
9. White RF, Proctor SP, Echeverria D, Schweikert J, Feldman RG. Neurobehavioral effects of acute and chronic mixed-solvent exposure in the screen printing industry. Am J Ind Med 1995; 28: 221–31.[Medline]
10. Parkinson DK, Bromet EJ, Cohen S, Dunn LO, Dew MA, Ryan C, Schwartz JE. Health effects of long-term solvent exposure among women in blue-collar occupations. Am J Ind Med 1990; 17: 661–75.[Medline]
11. DSM-IV: Diagnostic and statistical manual of mental disorders. 4th ed. Washington DC: American Psychiatric Association; 1994.
12. Hollingshead AB. Four-factor index of social status. New Haven (CT): AB Hollingshead; 1975.
13. Spitzer RL, Gibbon RL, Williams JBW. Structured Clinical Interview for DSM-IV axis I disorders (version 2.0). New York: Biometrics Research Department, New York State Psychiatric Institute; 1996.
14. Pilkonis PA, Heape CL, Ruddy J, Serrao P. Validity in the diagnosis of personality disorders: the use of the LEAD standard. J Consult Clin Psychol 1991; 3: 46–54.
15. First M, Spitzer RL, Gibbon M, Williams JBW. Computer assisted SCID–clinician version (CAS-CV). Washington DC: American Psychiatric Association; 1998.
16. Sanderson WC, Beck AT, Beck J. Syndrome comorbidity in patients with major depression or dysthymia: prevalence and temporal relationship. Am J Psychiatry 1990; 147: 1025–8.[Abstract/Free Full Text]
17. Reiger DA, Farmer ME, Rae DS, Locke BZ, Keith SJ, Judd LL, Goodwin EK. Comorbidity of mental disorders with alcohol and drug abuse: results from the epidemiologic catchment area (ECA) study. JAMA 1990; 264: 2511–8.[Abstract/Free Full Text]
18. Schottenfeld RS, Cullen MR. Organic affective illness associated with lead intoxication. Am J Psychiatry 1984; 141: 1423–6.[Abstract/Free Full Text]
19. Morrow LA, Ryan CM, Hodgson M, Robin N. Alterations in cognitive and psychological functioning after organic solvent exposure. J Occup Med 1990; 32: 444–50.[Medline]
20. Morrow L, Callender T, Lottenberg S, Buchsbaum M, Hodgson M, Robin N. PET and neurobehavioral evidence of tetrabromoethane encephalopathy. J Neuropsychiatry Clin Neurosci 1990; 2: 431–5.[Abstract/Free Full Text]
21. Royall DR. Frontal systems impairment in major depression. Semin Clin Neuropsychiatry 1999; 4: 13–23.[Medline]
22. Robinson RG, Starkstein SE. Mood disorder following stroke: new findings and future directions. J Geriatr Psychiatry 1989; 22: 1–15.
23. Starkstein SE, Robinson RG. Comparison of cortical and subcortical lesions in the production of poststroke disorders. Brain 1987; 110: 1045–59.[Abstract/Free Full Text]
24. Kessler RC, Zhao S. The prevalence of mental illness. In: Horwitz AV, Scheid TL, editors. A handbook for the study of mental health. Cambridge, UK: Cambridge University Press; 1999. p. 58–78.
25. National Institute for Occupational Safety and Health. Organic solvent neurotoxicity. Washington DC: US Government Printing Office; 1987. NIOSH Publication No.: 87-104.
26. Claxton AJ, Chawla AJ, Kennedy S. Absenteeism among employees treated for depression. J Occup Environ Med 1999; 41: 605–11.[Medline]
27. Bendiksen MS, Bendiksen I. A multi-dimensional intervention for a toxic solvent injured population. J Cogn Rehab 1992; 10: 20–6.
28. Haut MW, Leach S, Kuwabara H, Whyte S, Callahan T, Ducatman A, Lombardo L, Gupta N. Verbal working memory and solvent exposure: a position emission tomography study. Neuropsychology 2000; 14: 551–8.[Medline]

This article has been cited by other articles:

				G. T. Fernandes and M. G. Vaughn Brazilian street children: Contextual influences in relation to substance misuse International Social Work, September 1, 2008; 51(5): 669 - 681. [Abstract] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Morrow, L. A. Articles by Scott, A. Search for Related Content PubMed PubMed Citation Articles by Morrow, L. A. Articles by Scott, A. Related Collections Anxiety Depression Stress and Coping