Chronic drug use and cognitive impairments

Abstract

Reports regarding the effects of chronic drug use on human cognition and comparisons of different drug user groups have been inconsistent. Methodological flaws may account for some inconsistencies. To determine the effects of chronic drug use on cognition, drug users (n=192) were tested 17.1±0.3 days (mean±S.E.) and 99.4±1.7 days on average after their last use of drugs before beginning treatment. Drug users were categorized as stimulant, alcohol, or polydrug users. Their performance on tests of academic achievement, verbal memory, and abstraction was compared to performance of community-dwelling controls (n=137). The groups were matched on selected demographic and psychiatric characteristics. Historical records of achievement test scores were used to attempt to control for premorbid intellectual ability. Drug users showed impairments on each of the achievement tests (P<.001), as well as poorer total recall (P<.01) and abstraction ability (P<.05). Stimulant users performed worse on several tests relative to the other drug use groups. Only one of six tests demonstrated improvements following about 3 months of abstinence. Thus, chronic drug use is associated with cognitive impairments that do not improve substantially even after several months of abstinence.

Introduction

Studies of the chronic effects of drug abuse on cognition have produced inconsistent findings. Although certain studies have found deficits in memory, attention, abstraction, decision-making, and visuospatial abilities Bauer, 1996, Beatty et al., 1995, Beatty et al., 1997, Mittenberg and Motta, 1993, Rogers et al., 1999, Rosselli and Ardila, 1996, others have failed to find deficits in some of the same functions Bolla et al., 1999, Gillen et al., 1998, Hoff et al., 1996, O'Malley et al., 1992, Robinson et al., 1999. For example, contrasting findings that cocaine users performed poorer than, or equivalent to, controls have been reported on the Logical Memory subtest of the Wechsler Memory Scale Bolla et al., 1999, Rosselli and Ardila, 1996, the California Verbal Learning Test Gillen et al., 1998, Mittenberg and Motta, 1993, Trail Making Tests A and B Beatty et al., 1995, Bolla et al., 1999, O'Malley et al., 1992, and the Block Design subtest of the Wechsler Adult Intelligence Scale-Revised (WAIS-R) Beatty et al., 1995, Gillen et al., 1998, O'Malley et al., 1992.

Stimulant users actually performed better on certain measures than controls in a few studies Bolla et al., 1999, Hoff et al., 1996, Manschreck et al., 1990, McKetin and Mattick, 1998, O'Malley et al., 1992. Cocaine users, relative to controls, completed mazes more quickly (Manschreck et al., 1990) and performed better on the Rey Auditory Verbal Learning Test, Trial 1 (Bolla et al., 1999), the Oral Fluency part of an abbreviated version of the Benton Multilingual Aphasia Exam (O'Malley et al., 1992), and the number of correct categories on the Wisconsin Card Sorting Test and the Controlled Oral Word Association Test (Hoff et al., 1996). Amphetamine users who were assessed as being in a low-dependence group performed better than controls on the verbal memory index of the Wechsler Memory Scale-Revised (McKetin and Mattick, 1998).

Studies comparing the severity of cognitive deficits associated with different drugs have also shown discrepant results. One study (Selby and Azrin, 1998) found that cocaine users performed better than alcohol and polydrug users, whereas several other studies found that cocaine users generally performed worse than alcohol and polydrug users Bauer, 1994, Bauer, 1996, Beatty et al., 1995, Robinson et al., 1999, Rosselli and Ardila, 1996.

Methodological flaws in studies may account for some of the inconsistencies mentioned above. Virtually no studies have included measures of premorbid cognitive function, raising the possibility that differences between drug users and controls, or differences among classes of drug users, existed before the onset of drug use, rather than being caused by drugs. Although a few studies attempted to estimate premorbid cognitive function Beatty et al., 1995, Berry et al., 1993, McKetin and Mattick, 1997, McKetin and Mattick, 1998, McKetin and Solowij, 1999, such estimates have limited validity. A number of studies have not included a control group of nonusers Ardila et al., 1991, Dafters et al., 1999, Horner, 1997, Krystal and Price, 1992, McKetin and Mattick, 1997, Strickland et al., 1993. To convincingly demonstrate cognitive deficits in drug users, comparison with an appropriately matched control group is essential. Many studies have involved small sample sizes, for example, as few as eight (Strickland et al., 1993) drug users. In the present study, we avoided these methodological flaws by retrieving historical records of achievement test scores to measure premorbid cognitive function, including a control group, and using a large sample size.

We compared three primary categories of drug users—stimulant, alcohol, and polydrug users. These broad categories provided sufficiently large group sizes to afford adequate power. We also conducted more fine-grained analyses, for example, comparing cocaine and amphetamine users within the stimulant group. These more fine-grained analyses provided justification for the pooling of subjects into the primary categories, as did other considerations, for example, with respect to cocaine and amphetamine, both drugs stimulate the central nervous system at many levels, have prominent effects on dopamine and other monoaminergic neurotransmitters, and produce similar patterns of intoxication and clinical problems (Schuckit, 2000, pp. 120–121).