We know little about substance use among patients with comorbid substance-related disorder (SRD) and dysthymia (1-3). This paucity of understanding probably stems from two issues: (a) the long duration required for a dysthymia diagnosis (i.e., two years) and (b) minor depressive symptoms that resemble uncomplicated early sobriety (4). Dysthymia could be associated with the type of substance abused, as has been observed in comorbid anxiety disorder and SRD (5). Although dysthymic patients may preferentially use particular substances to "treat" their symptoms, Leibenluft and coworkers found that alcoholics with and without depression used alcohol to treat depressive symptoms to the same degree (6).
Several advantages might accrue if we knew more about the substances that the dysthymic group used and abused. First, we might learn whether patients with dysthymia are using self-treatment (e.g., with stimulants for fatigue, sedatives for insomnia). Second, if patients with dysthymia are especially apt to abuse one or a few substances, we might be better able to recognize comorbid dysthymia earlier in recovery rather than waiting for a prolonged period (e.g., recognize certain drug use as a risk factor for dysthymia). Third, the substance use pattern over time might help us to understand better how substance abuse develops as a comorbid condition for dysthymia or vice versa. For example, individuals with dysthymia may decide not to use a given substance in an ongoing fashion even though they have had experience with that substance. Presumably, a choice not to continue use could be due to untoward effects during intoxication, untoward effects postintoxication, or untoward effects at both times. Alternatively, patients with dysthymia may use illicit drugs at a rate different from patients with SRD only; thus, differences in exposure rates may affect prevalence rates of particular SRD diagnoses.
To address these issues, we compared a group of patients with comorbid SRD-dysthymia and SRD only in relation to the following:
* lifetime use of the substance
* at first use of the substance
* years of use of the substance
* number of days the substance was used in the last year
* SRD diagnoses
A total of 642 patients was studied at two alcohol-drug programs located in psychiatry departments at two university settings (Minnesota and Oklahoma). All patients with both SRD and dysthymia (n = 39) and all patients with SRD only (n = 308) were selected for the study. Patients with so-called dual depression (dysthymia + major depressive disorder) were excluded from this analysis. All data included in this study had been collected using standard instruments for purposes of research, as well as assessment and treatment planning. These included the following:
* self-rated instruments: 90-item Symptoms Checklist (7) and Beck Depression Inventory (BDI) (8)
* a trained research associate obtained detailed demographic information and substance use history using a specific questionnaire format, as described in an earlier report (9)
* psychiatrist-rated instruments: Hamilton Depression Scale (HAM-D) (10) and the Brief Psychiatric Rating Scale (11)
On entry into the alcohol-drug program for assessment, the patients provided informed consent for the scientific use of these data in an anonymous and confidential fashion.
Trained research associates obtained the following information: types of substances ever used, age at first use, number of days of use in the last year, and number of years in which any use occurred. In a small number of cases, the subjects could not recall specifically the age at first use. In these cases, the research associate determined the age at first use based on the best estimate from data available; there were no missing values for these data. The years of substance use were determined by subtracting from the patient's current age the age at first use plus periods of abstinence that lasted months to years; there were no "zero values" for these data. Patients with elevated scores on the BDI or HAM-D were monitored with monthly BDI reassessments until the BDI fell to normal limits or until an appropriate diagnosis was made; this process continued over the first several months as appropriate to the individual patient.
A psychiatrist specializing in addiction made the current SRD diagnosis at the time of the initial evaluation, as well as the dysthymia diagnosis. Most cases of dysthymia had a prior history of dysthymia during 2 or more previous years of continuous sobriety; in these cases, we observed the patient for an additional 6 months to confirm the presence of chronic minor depressive symptoms consistent with dysthymia. In a small number of cases, the patients had been continuously abstinent during the previous 2 years and had continued to manifest chronic minor depressive symptoms despite abstinence and other efforts at recovery (e.g., Alcoholic Anonymous meetings, care from a counselor for chemical dependency); these patients received dysthymia diagnoses immediately on initial evaluation. Strict criteria for tobacco dependence were employed (i.e., an identified health problem, professionally prescribed abstinence, and failure to achieve abstinence).
The two groups were compared using the following statistical measures:
* Chi square with continuity correction for nonparametric data (e.g., lifetime drug use, SRD diagnosis)
* Mann-Whitney test for nonparametric data (e.g., age at first use, years of use)
Level of statistical significance was set at .01 in view of the large number of comparisons. Significance levels of .02 to .05 were reported as borderline trends.
The patients with SRD-dysthymia were more apt to be currently living with their parents and less apt to be living with a spouse (P [is less than] .01), although they did not differ notably for other types of residence (i.e., alone, with friends, halfway house and other institutions, etc.). The two groups did not differ regarding age, sex ratio, education, marital status, or employment status.
Lifetime Use of Psychoactive Substances
As shown in Table 1, the two groups had highly similar lifetime use (i.e., any use during one's entire previous lifetime, or lifetime prevalence) of psychoactive substances. Four substances showed almost identical life use prevalence: caffeine, alcohol, amphetamines, and phencyclidine (PCP).
Table 1. Substance-Related Disorder-Dysthymia Versus Substance-Related Disorder Only
(a) Number of subjects in one or more cells is 5 or less; Fisher Exact Chi Square test was performed.
Age at First Use
The chronological order for initiating substance use was approximately the same for the patients with SRD-dysthymia and the other patients (Table 1). They began with legal or highly available substances during early adolescence (i.e., caffeine, inhalants, alcohol, tobacco), progressed to illicit "soft" drugs during middle adolescence (i.e., hallucinogens, PCP, cannabis, amphetamines, sedatives), and then progressed to the "hard" drugs in late adolescence (i.e., opioids and cocaine). However, the group with SRD-dysthymia reported beginning caffeine use significantly earlier than the other group (P [is less than] .005). Differences in first use for the remaining 10 substances were not statistically significant. The group with SRD-dysthymia started using 5 of these 10 substances at an earlier mean age, and the other group started using 5 different substances at an earlier mean age.
Years of Use
The years of use variable (Table 1) consisted of the number of years in which the psychoactive substance was used. This duration of substance use was not significantly different for those substances used over long periods of time (e.g., caffeine, tobacco, alcohol, cannabis). However, there were significant differences for opioids and a borderline difference for cocaine and amphetamines. For each of these three substances, the patients with SRD-dysthymia used them for only a few years on average, whereas the group with SRD only tended to use these drugs for several years. The group with SRD-dysthymia had a shorter duration of use for 8 of the 11 substances. The duration of caffeine use was unexpectedly low for the group with SRD-dysthymia in view of their early onset of use. Many patients in both groups reported intermittent use or discontinuation of caffeine use due to problems with taking excessive amounts (e.g., caffeinism) or associated medical problems (e.g., headache, bowel problems, peptic ulcer).
Days of Use in the Last Year
Both groups had similar frequency of use during the last year for the most commonly used substances (i.e., tobacco, alcohol, cannabis) (Table 1). However, the group with SRD-dysthymia used cannabis and cocaine on significantly fewer days (P [is less than] .004, P [is less than] .01, respectively). The patients with SRD-dysthymia also used opioids and amphetamines less often, but only to a borderline extent (P = .03, P = .02, respectively).
Substance-Related Disorder Diagnoses
As shown in Table 1, the group with SRD-dysthymia was significantly less apt to meet criteria for a cannabis abuse/dependence diagnosis (P [is less than] .001). The rate of tobacco dependence was low in both groups. A few adults continued to abuse inhalants, usually episodically or in special circumstances; especially common was the use of amyl nitrate capsules to delay orgasm.
If a difference were to occur in duration and frequency of use, one might have expected that those with dysthymia would have heavier use of stimulant drugs such as cocaine and amphetamines. On the contrary, the patients with dysthymia had used both substances less often in the last year, and they had used cocaine for fewer years. These findings were not due to reduced exposure to these substances since the lifetime use of these substances did not differ between the two groups. Of possible relevance, Castaneda and coworkers found that cocaine users reported a worsening of their psychiatric symptoms with cocaine use (12). If the findings of Castaneda and coworkers apply to patients with SRD-dysthymia, then these patients may cease abusing these substances because they experience more severe dysthymia in association with use of stimulants. This could also apply to caffeine since patients with SRD-dysthymia had fewer lifetime years of caffeine use than one might have anticipated given their early onset of caffeine use. They began use of caffeine significantly earlier than patients with SRD only (P [is less than] .005). This unexpected finding could be an early attempt at self-treatment if mood changes started early in childhood. Such mood changes have been shown to pre-date, or at least accompany, drug abuse and dependence in adolescents (13-15). Although early caffeine use does not seem likely as a cause of later dysthymia, the possible induction of dysthymia from heavy early caffeine use should not be discounted out of hand (16). Pardo, Pardo, and Raichle have recently found inferior and orbitofrontal cortical activation in association with dysphoric feelings in normal individuals (17), suggesting a discrete neural substrate for such feelings. Alternatively, this finding of early caffeine use in SRD-dysthymia may be spurious since it is a solitary finding (unlike other significant findings in this study, which support each other). Additional studies should be undertaken to assess the replicability of this suggestive finding.
Use of cannabis began a few years later on average among the patients with SRD-dysthymia, but the difference was not significant. Years of use did not differ notably between the two groups, but patients with SRD-dysthymia had significantly less use during the last year (P [is less than] .004) Diagnoses of cannabis abuse/dependence were significantly fewer in the group with SRD-dysthymia (P [is less than] .001). It appears that, like cocaine, cannabis may make dysthymia worse. Castaneda et al.'s data would suggest that this worsening may occur in the postintoxication period (12).
Use of opioids was also different between the groups, with patients with SRD-dysthymia using for a significantly shorter period (P [is less than] .01). Their frequency of use over the last year showed a trend for fewer days of use (P = .03). These data suggest, but do not prove, that patients with SRD-dysthymia may obtain slightly less desirable effects from opioids as compared with patients with SRD only.
Alcohol history and diagnoses were remarkably similar in both groups. They had the same lifetime use, similar age at onset of use, equal years of use, almost the same number of days of drinking during the previous year, and comparable diagnostic rates of alcohol abuse/dependence. The pattern of inhalant and hallucinogen use did not show any statistically significant difference between the two groups. There were slight trends for the group with SRD-dysthymia to have used these drugs on somewhat fewer days during the last year. Dysthymia appeared to auger neither for nor against alcohol, inhalants, or hallucinogens.
In summary, patients with SRD-dysthymia may have begun using caffeine during childhood as an effort toward self-treatment. Subsequently, patients with SRD-dysthymia manifested lower indices of stimulant, opiate, sedative, and cannabis use/abuse. It seems likely that prolonged or recurrent use of these substances worsened mood symptoms in SRD-dysthymic patients.
Partial support for this project was obtained from the Laureate Foundation of Tulsa. Dr. James Halikas, Mr. John Neider, and Mr. Greg Carlson collaborated in the collection of these data.
(1.) Kell, M. J., Opiate dependence, comorbidity and seasonality of birth. J. Addict. Dis. 14(3):19-34 (1995).
(2.) Keller, M. B., Dysthymia in clinical practice: Course. outcome and impact on the community, Acta Psychiatr. Scand. 383(Supplementum):24-34 (1994).
(3.) King, C. A., Naylor, M. W., Hill, E. M., et al., Dysthymia characteristics of heavy alcohol use in depressed adolescents, Biol. Psychiatry 33(3):210-212 (1993).
(4.) Powell, B. J., Penick. E. C., Nickel, E. J., et al., Outcomes of co-morbid alcoholic men: A 1-year follow-up, Alcohol. Clin. Exp. Res. 16(1):131-138 (1992).
(5.) Tucker, P., and Westermeyer, J., Substance abuse in patients with comorbid anxiety disorder, Am. J. Addict. 4(3):226-233 (1995).
(6.) Leibenluft, E., Fiero, P. L., Bartko, J. J., et al., Depressive symptoms and the reported use of alcohol, caffeine, and carbohydrates in normal volunteers and four groups of psychiatric outpatients, Am. J. Psychiatry 150(2):294-301 (1993).
(7.) Derogatis, L. R., Lipman, R. S., and Covi, L., The SCL-90: An outpatient psychiatric rating scale, Psychopharmacol. Bull. 9:13-28 (1973).
(8.) Beck, A. T., Ward, C. H., Mendelson, M., et al., An inventory for measuring depression, Arch. Gen. Psychiatry 4:561-571 (1961).
(9.) Westermeyer, J., and Eames, S. E., Clinical epidemiology of comorbid dysthymia and substance disorder, Am. J. Addict. 6(1):48-53 (1997).
(10.) Hamilton, M., A rating scale for depression, J. Neurol. Neurosurg. Psychiatry 25:56-61 (1960).
(11.) Overall, J. E., and Gorham, D. R., Brief Psychiatric Rating Scale, Psychol. Rep. 10:799-812 (1962).
(12.) Castaneda, R., Lifshutz, H., Galanter, M., et al., Empirical assessment of the self-medication hypothesis among dually diagnosed inpatients, Psychiatry 35(3):180-184 (1994).
(13.) Burke, J. D., Burke, K. C., and Rae, D. S., Increased rates of drug abuse and dependence after onset of mood or anxiety disorders in adolescence, Hosp. Community Psychiatry 45(5):451-455 (1994).
(14.) Deykin, E. Y., Buka, S. L., and Zeena, T. H., Depressive illness among chemically dependent adolescents, Am. J. Psychiatry 149(10):1341-1347 (1992).
(15.) Weiss, R. D., Griffin, M. L., and Mirin, S. M., Drug abuse as self-medication for depression: An empirical study, Am. J. Drug Alcohol Abuse 18(2):121-129(1992).
(16.) Extein, I. L., and Gold, M. S., Hypothesized neurochemical models for psychiatric syndromes in alcohol and drug dependence, J. Addict. Dis. 12(3):29-43 (1993).
(17.) Pardo, J. V., Pardo, P. J., and Raichle, M. E., Neural correlates of self-induced dysphoria, Am. J. Psychiatry 150(5):713-719 (1993).
COPYRIGHT 1998 Marcel Dekker, Inc.
COPYRIGHT 2001 Gale Group