Brief screening for cognitive impairment in addictive disorders.

Chronic use of mind altering substances can lead to a wide variety of neuropsychological deficits, affecting the domains of attention, learning, memory, reasoning. Executive functions such as working memory, cognitive flexibility and inhibitory control may specifically be impaired. These deficits can impact engagement in effective psychosocial interventions. Mild to moderate cognitive dysfunction may not be picked up in routine clinical examination or through commonly used tests like the mini-mental state examination (MMSE). Detailed neuropsychological tests, although extremely valuable, are time and human-resource intensive and are not readily available to the clinician. This study attempted to devise a brief cognitive screen (BCS- AUD) for alcohol use disorders. Ninety subjects who fulfilled ICD-10 criteria for alcohol use disorders were assessed on the MMSE and selective tests from the NIMHANS neuropsychological battery. While 79 (87.78%) of patients had adequate scores on the MMSE (>25), cognitive deficits were noted with relatively high frequency on finger tapping (92.22-93.33%), auditory verbal learning test delayed recall AVLTDR (37-63%) and Tower of London 5 move subtest (42%). Statistically significant associations were found between MMSE and Digit symbol total time (0.05), Finger tapping right hand (0.01), Tower of London total number of problems solved with minimum moves (TNPSMM) (0.05), Verbal working memory two back hits (VM2BKHIT) (0.01), AVLTDR (0.01), and complex figure test-copy (0.01). Principal component analysis helped to identify three tests that merited inclusion in the BCS-AUD, namely Finger Tapping Test, Verbal Working Memory N Back Test and Auditory Verbal Test (AVLT). The utility of the BCS-AUD in identifying cognitive dysfunction in other substance use disorders needs to be examined. Patients rating positive on the cognitive screener would require in-depth evaluation, monitoring and remediation.

INTRODUCTION

In 1901, Bonhoeffer demonstrated memory dysfunction in cases of delirium tremens.1 Since then, there have been a series of studies examining cognitive dysfunction in alcoholism, ranging from evaluating intelligence quotients (IQ)2 to evaluating more subtle derangement of cognitive functioning.345 Alcohol tends to affect elective mental capacities, rather than having a diffuse impact on mental function.

The brain regions and neural processes underlying addiction overlap extensively with those that support cognitive functioning, including learning, memory and reasoning.6

Cognitive deficits associated with alcohol use

A large number of neuropsychological deficits are associated with chronic alcohol use. Investigations in alcohol dependent individuals indicates diffuse deficits involving the areas of attention, delayed response ability, psychomotor functioning, ideational fluency, abstraction, problem solving, visuo-spatial functions, visual integration, verbal and visual learning and memory functions.7 Specific deficits in coding and retrieval were found in male alcohol dependent patients as compared to social drinkers.8 In a group of 30 patients with alcohol dependence who underwent a detailed neuropsychological assessment at the Centre for Addiction Medicine at NIMHANS, a range of deficits were found in response inhibition (55.9%), conceptual responses on the Wisconsin Card Sorting Test (50%), phonemic fluency and visual working memory (29.4%), visual and verbal 1 back task and failure to maintain set (26.5%) conceptual level responses and visual 1 back hits (23.5%), animal fluency and on verbal 2 back errors (20.6%).9

Cognitive deficits associated with other substance use

Cognitive deficits have also been shown to be present in other chronic drug use, including deficits in cognitive flexibility in cocaine10 and opioid users11, deficits in attention and impulse control in amphetamine users12, deficits in cognitive flexibility and attention in cannabis users and deficits in working memory and declarative learning in tobacco smokers.13

Prevalence

Estimates regarding the prevalence of cognitive impairment in SUD patients vary widely and range from about 30-80%.14

Nature of cognitive deficits

Tarter15 classifies neuropsychological deficits as either antedating the onset of drinking, concurrent to alcohol use or consequential to alcoholism, with a higher occurrence among children with perinatal injury, poor memory and visuospatial integration, lower IQ, exposure to physical abuse. Cognitive and behaviour deficits in children exposed in utero to alcohol1617 cannabis18 methamphetamine1920 are well recognised.

Impact of cognitive deficits

Cognitive deficits have an impact on treatment and outcome, including patient's understanding of the problem, as well as with treatment compliance and relapse. These cognitive deficits may be amenable to retraining, thereby having an impact on the overall prognosis.

A knowledge of past and present cognitive dysfunction is important in treating persons with addiction, as these cognitive changes may result in what is perceived as ‘denial’ and the person continues to engage in behaviours that maintain the addictions. Such cognitive deficits may also hinder the person's ability to benefit from counselling and more sessions and/or reminders may be necessary to aid these patients in incorporating abstinence-sustaining strategies into their daily routines.

Transient or Persistent?

Acute intoxication and immediate and protracted withdrawal produce transient alterations of cognitions that can persist for weeks to months. Some subtle residual effects remain for up to one year for certain drugs.21

While the cognitive deficits associated with withdrawal from drugs are often temporary, long-term use can also lead to lasting cognitive decline.6 The nature of the deficits varies with the specific drug, the environment and the user's genetic make-up.6 All the deficits are particularly pronounced during early periods of abstinence.

It is of clinical importance to understand whether deficits are reversible with abstinence22 and the possible relationships of cognitive dysfunction to psychosocial functioning. Chronic ingestion of alcohol triggers deficits in certain specific neuropsychological functions that are linked with structural and functional disorders of the anterior frontal region, which can be recovered to a large extent when consumption ends.23 A follow-up study of 312 subjects with alcohol dependence at NIMHANS showed that the cognitive impairment could be significantly reversed through cognitive retraining.24 Retraining was found to help in a significant reduction in the number of neuropsychological deficits in the area of working memory, as well as visual learning and memory functions.25

Some recent studies suggest that some of the methamphetamine induced cognitive losses may be partially recouped with extended abstinence2627 while other suggest that certain cognitive deficits may remain even after 2.5 years of abstinence.28 In a study among smokers, cognitive speed declined nearly twice as much as non-smokers over 5 years and declines in smokers ‘cognitive flexibility and global cognition occurred at 2.4 times and 1.7 times the respective rates of nonsmokers.29

Assessment of cognitive dysfunction and need for a brief cognitive screener

While it is easy to identify cognitive deficits in persons with alcohol-induced amnestic disorders, oftentimes, the cognitive deficits are too subtle to be picked up in routine evaluation in patients with substance use disorders presenting in uncomplicated withdrawal. At the present time, the available of cognitive evaluations is through either standardized intelligence tests or through extensive neuropsychological test batteries. It becomes necessary in clinical practice to develop a simple battery of tests which can be applied in clinical practice.

An extensive neuropsychological battery is used at NIMHANS to assess cognitive dysfunction in a variety of conditions. This consists of 21 subtests assessing different domains, is time consuming and requires specialised training. This article describes the development of a brief cognitive screening instrument for cognitive dysfunction in persons with alcohol dependence, which can be used in clinical settings.

METHODOLOGY

Ninety male inpatients between the age group of 20-60 years, satisfying ICD 10 criteria for alcohol dependence, were assessed 2 weeks after completion of detoxification, or 3 weeks after the last drink, whichever was earlier. Other inclusion criteria included a minimum education upto class one and the ability to cooperate for neuropsychological assessment. Patients with independent psychiatric or major physical illnesses, concurrent substance related disorders apart from nicotine, patients with an organic brain syndrome, seizures, delirium tremens and Wernicke Korsakoff's syndrome were excluded. Participants were administered a socio-demographic questionnaire, the Severity of alcohol dependence questionnaire (SADQ), Mini mental state examination30 and the NIMHANS neuropsychological battery.31 From the NIMHANS neuropsychological battery, tests selected included the Finger tapping test32, Category fluency33, Verbal working memory N Back test34, Tower of London test35, Rey's auditory verbal test36 digit symbol substitution test,33 digit vigilance test33 and the complex figure test. The tests were selected based on the domains likely to be impaired in persons with chronic alcohol use, based on a review of literature.

Abstinent inpatients were administered the tests ensuring enough time in between tests to avoid fatigue. The tests were carried out over a duration of 1-1.5 hours, over 2-3 sessions to prevent fatigue. For the statistical analysis, mean scores on neuropsychological and other clinical variables were compared using t test and ANOVA. The relationship between socio-demographic and neuropsychological variables were determined by Pearson correlation coefficient. An inter-correlation was done in order to understand the relationship across the domains measured by the tests. The data obtained was compared to a normative cut-off score, provided in the NIMHANS neuropsychology manual, to determine deficits in respective areas. Deficit frequencies were compared between the MMSE and the neuropsychology variables. Tests which explained the maximum variance in the total set of variables were chosen, by subjecting the scores to a factor analysis. Thus a set of 8 neuropsychological variables were selected for factor analysis and based on sensitivity and factor analysis, 3 tests were selected for the neuropsychology battery.

Sample characteristics:

Mean age of the sample was 38.14 (8.45) years. A majority were married (84.4%), residing in an urban locality (73.3%) and employed (73.3%). While 5 (5.6%) patients were functionally literate, levels of education varied with 17 (18.9%) having received 1-5 years of formal education, 46 (51.5%) 6-12 years of formal education, 9 (10%) higher secondary education and 13 (14.5%) were graduates. A majority of them consumed whisky (84.4%). Seven patients (7.8%) reported a history of consuming licit arrack. Age at initiation of alcohol was 23.93 (5.5) years and mean duration of drinking was 14.21 (7.59) years. Maximum units of alcohol per day was 15.96 (6.99). Mean SADQ score was 33.20 (8.11) suggesting severe dependence. The average score on MMSE was 27, above the cut-off of 25, which suggests impairment. 79 subjects had scores above 25 on the MMSE.

Neuropsychological variables

Table 1 and Figure 1 show the number and percentage of patients with neuropsychological deficits and upto 20% of patients with normal MMSE had a range of deficits on the tests administered.

Table 1

Number and percentage of patients who showed neuropsychological deficits on testing

Figure 1

Pattern of deficits on different neuropsychological variables

In summary, deficits were found in digit symbol substitution time (20%), verbal 1 Back hit (25.56%), verbal 2 Back hit (51.1%), finger tapping 92.22-93.33%), category fluency (16.67%), complex figure test copy (23%), immediate recall (18.9%), delayed recall (23.33%), auditory verbal learning test AVLTDR (37-63%), Tower of London 5 move subtest (42%) [Figure 1].

The deficit frequency on MMSE scores and neuropsychology variables were compared. Statistically significant associations were found between MMSE and Digit symbol total time (0.05), Finger tapping right hand (0.01), TNPSMM (0.05), VM2BKHIT (0.01), AVLTDR (0.01), and complex figure test-copy (0.01) [Figure 2].

Figure 2

Comparisons on Deficit Frequency between Adequate MMSE and Impaired Neuropsychology Variables

Three factors (finger tapping, AVLDTR, VM2BKHIT) were loaded for the principal component factor analysis, with orthogonal rotation and factors with Eigen values > 1 were selected. Variable with factor loading of 30 were taken to explain the factors. Together, the 3 factors explained 57% of the variance.

The accompanying component plot Figure 3 shows NP variables as loaded on rotated component matrix. Variables mentioned in bold and arrows were taken for the neuropsychology battery.

Figure 3

Component Plot- variances on rotated component matrix

The Brief Cognitive Screen for Alcohol use disorders (BCS-AUD) recommended from the findings include the following:

Finger tapping Test32

Verbal Working Memory N Back Test34

Rey's Auditory Verbal Test (AVLT)36

These 3 tests should take about 25-35 minutes.

DISCUSSION

Cognitive deficits are known to have an impact on a variety of outcomes in substance use disorders. These include poor treatment retention and lower levels of abstinence.37 The reason for this is that cognitive dysfunction may interfere with the development of insight and association with denial,38 factors which might interfere with the ability of patients to be engaged in psychosocial interventions.

There is a need to develop a routine screening instrument for cognitive function in persons with substance use disorders and monitor cognitive functions at follow-up, both to determine the restitution of these functions with abstinence, as well as to monitor for lasting cognitive impairment, which may warrant the use of other psychosocial strategies apart from those routinely used in addiction recovery.

There is also a need to develop a brief cognitive screen in health personnel strained resource settings. This may help to pick up subtle cognitive deficits which may not be picked up in tests like the MMSE.

The present study derived a small sub-set of tests from the detailed neuropsychological battery, based on the deficits commonly noted in persons with alcohol dependence. This screener could alert the clinician to identify subtle cognitive deficits in patients, arrange for a detailed cognitive evaluation and remediation and guide psychosocial intervention in an individual patient. It needs to be determined in further studies whether the same parameters would be useful for cognitive screening for patients using other substances.

In more recent years, the need for rapid screening of patients with substance use disorders has gained greater attention, and tools like the Montreal Cognitive Assessment (MoCA), which was originally developed for a geriatric population to identify mild cognitive impairment have been evaluated in patients with substance use disorders,14 with mild-moderate impairment. This 10-minute tool showed acceptable sensitivity (83.3%) and specificity (72.9%) for the identification of cognitive impairment (with a cut-off score of 25) when compared to a standardised 45 minute battery.

The MoCA includes tests of attention (Orientation, digits forward, digits backward, serial sevens), Language (picture naming), memory for words (immediate and delayed recall and sentence repetition), spatial (cube drawing and clock drawing) and executive (alternating trail making, verbal fluency through word generation) and abstraction. While the BCS-AUD has been specifically designed for alcohol use disorder populations, the MoCA has developed for dementia. The attractiveness of the latter is the shorter time.

In future studies, we hope to compare the BCS-AUD with MoCA in persons with alcohol use disorder and examine its usefulness as a screener in patients with other substance use disorders to screen for cognitive dysfunction for appropriate management and remediation.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.