Externalizing psychopathology and cognitive functions in patients with early- and late-onset alcohol dependence.

BACKGROUND: Alcohol use disorder is attributing to a significant health-care burden worldwide. Early-onset alcohol dependence is associated with more adverse outcomes than those with late-onset alcohol dependence. Comorbid externalizing disorders and cognitive deficits may be associated with the negative outcomes in early-onset alcohol dependence. This study aims at exploring the externalizing psychopathology and cognitive performance in early-onset alcohol dependence versus late-onset alcohol dependence.
MATERIALS AND METHODS: This is a cross-sectional study carried out on patients attending the psychiatry unit of a tertiary care center of north India after obtaining approval from the institutional ethics committee. A total of 57 patients with alcohol dependence enrolled in the study, after screening a total of 112 patients. Patients were evaluated for the externalizing psychopathology (using SSAGA intravenous [IV]) and cognitive performance (using Wisconsin Card Sorting Test [WCST] and continuous performance test [CPT]). Comparison of sociodemographic, clinical variables as well as externalizing psychopathology and cognitive performance was done between early-onset and late-onset alcohol dependence.
RESULTS: Comparison between early-onset and late-onset alcohol dependence revealed that the score of individual externalizing psychopathologies and the total externalizing psychopathology score on SSAGA IV in the early-onset group are significantly higher than late-onset alcohol dependence. Similarly, there is a significant difference in the executive functions (on WCST) between the two groups (early onset < late onset). On CPT, there are significantly more errors of omission in the early-onset group in comparison to their late-onset counterparts.
CONCLUSION: Early-onset alcohol dependence is associated with more externalizing psychopathology and more cognitive dysfunction than late-onset alcohol dependence. Copyright:

INTRODUCTION

Alcohol use disorder (AUD) is a global problem and has a major contribution to both morbidity and mortality. It encompasses a wide range of disabilities both psychiatric and medical. It has a direct health impact on the user such as alcohol dependence or liver cirrhosis or indirect effect on others via the potentially dangerous acts of a person in the state of intoxication. As per the report of global burden of diseases, AUD is the most common substance use disorder worldwide, with an estimate of 100.4 million cases in 2016.[1] Approximately 2.5 million deaths occur every year due to the use of alcohol, globally. AUD is responsible for 5.1% of the disability-adjusted life years.[2] The National Mental Health Survey 2015–2016 reports that the prevalence of AUD in community representative population is 4.65%.[3]

Age of onset of alcoholism plays a pivotal role in the pathogenesis of AUD. The phenomenology, course, outcome of early-onset, and late-onset AUD are different. However, there is a lack of consensus on the age demarcation that differentiates early-onset from late-onset AUD. Many studies use the age cutoff ranging between 20 and 25 years as the demarcation and most of them agree upon 25 years as the demarcation between early-onset and late-onset AUD.[4567]

”Externalising symptoms” refers to a group of behavioral problems manifested as the outward behavior and negatively reacting to the external environment. They consist of hyperactive, disruptive, and aggressive behaviors which are generally evident in childhood and can also persist till adulthood.[8] The spectrum of externalizing disorders consists of childhood disruptive disorders (oppositional defiant disorder (ODD), conduct disorder [CD]), attention-deficit hyperactivity disorder (ADHD), antisocial behavior, and personality. Various studies indicate that there is extensive comorbidity of these disorders with substance use disorders.[910] It has also been seen that externalizing psychopathology can robustly predict the early onset of substance use disorder. Thus, early disinhibition of behavior may be followed by early substance use and these set of findings are thought to be of common origin.[11] It has been recently observed that in India that the age of first using alcohol has gone down from an age of 28 years (in the 1980s) to 17 years (in 2007).[12] Taking into account the drop in the age of first alcohol use in recent trends in India, a study focusing on the age of onset was inevitable for risk stratification necessary mitigation.

In the field of cognition, executive functions form an important component that encompasses self-regulatory functions such as planning, cognitive flexibility, working memory, sense of time, and inhibition. It has been seen in various studies that patients with AUD have a poor performance in cognitive executive functions.[13] In substance dependence, there is diminished performance in executive functioning as well as abnormality in related brain structures.[14] On the other hand, the executive function deficit can also increase the risk of developing substance use disorders.[15] Specific studies on alcohol also showed impairment in a wide range of executive domains. Alcohol has effects on some components of executive functioning which include updating, set-shifting, and response inhibition.[15] There is a scarcity of literature to see the differences in cognitive functions if any in early- and late-onset alcohol dependence. This is a part of a larger study, which aimed to evaluate the psychiatric comorbidities, the severity of the addiction, externalizing psychopathology, and cognitive performance in patients with early and late-onset alcohol dependence attending a tertiary care teaching hospital of North India. In the current paper, we are presenting the data that focused on the association of externalizing psychopathology and cognitive performance in patients with early and late-onset alcohol dependence. We hypothesized that the patients with early-onset alcohol dependence will have higher externalizing psychopathology scores and worse performance in the cognitive tests even after similar educational profile, duration of illness, and prescribing them the same medications, in comparison to those with late-onset alcohol dependence.

METHODOLOGY

Settings

This is a cross-sectional study conducted on patients with AUD attending the psychiatric unit of a tertiary care center in North India.

Inclusion and exclusion criteria

Patients fulfilling the diagnosis of alcohol dependence as per the International Classification of Disease-10 (ICD-10) Diagnostic Criteria for Research (DCR) criteria, willing to give written informed consent, and accompanied by relatives who could corroborate with the history provided by the patient were included in the study. Subjects with serious medical comorbidities or too ill to co-operate for the interview were excluded from the study. Subjects with formal schooling <8 years, subjects with the clinical picture of subnormal intelligence, and visual- and hearing-deficit patients were excluded so as not to influence the performance of the computerized cognitive tests.

Assessment tools

The diagnosis of the patients was made following the ICD-10 DCR diagnostic criteria. All diagnosed patients of alcohol dependence were assessed on a semi-structured pro forma for their sociodemographic information and alcohol use history. Semi-structured assessment for the genetics of alcoholism (SSAGA) IV was used to estimate the age of onset (by its lifetime interview section), externalizing psychopathology (by adding up the symptom domains under the sections of CD, dissocial personality disorder, ADHD Inattention, ADHD Hyperactivity, ODD any 6 months, ODD worst 6 months).

Wisconsin Card Sorting Test (WCST) (for executive function) and the continuous performance test (CPT) (for attention and concentration) were applied to the patients to find out their cognitive performance.

Procedure

The study was conducted after obtaining ethical clearance from the institutional ethics committee. Informed consent was obtained from the patients. Patients diagnosed with alcohol dependence, attending the department of psychiatry, and meeting the selection criteria of the study were evaluated. Patients with a diagnosis of other alcohol-related disorders (intoxication, withdrawal, or psychosis) were followed up regularly until their condition became clinically stable. If they qualified for recruitment in follow-ups as per the selection criteria of the study, then they were included.

Patients were assessed on semi-structured pro forma. Age of onset and externalizing psychopathology were assessed on SSAGA IV. History and relevant information were confirmed from the source of a reliable informant [Figure 1]. The operational definition was taken for “age of onset”-the age at which the patient first fulfilled the diagnosis of alcohol dependence (as per self-report and structured interview on SSAGA IV).

Figure 1

Flowchart showing the methodology of the study

Externalizing symptom score was calculated by taking each point of the domains as mentioned before and a total score was obtained after adding up the individual scores. The patients were started on appropriate medications (e.g., thiamine, benzodiazepines, zolpidem, and trazodone) and cognition was checked by the WCST (128 card version) and CPT (Conner's) after at least 2 months. The period of 2 months was kept for various reasons. First, autonomic hyperarousal due to acute withdrawal may lead to false results. It has been observed that the symptoms of even protracted withdrawal subside by 8 weeks. Second, a high dosage of benzodiazepines in the initial phase of treatment may affect results – convention and clinical practice shows that benzodiazepines are tapered down in 8 weeks. Third, effect due to state factors such as the heavy usage of alcohol and symptoms of untreated co-morbidities can be mostly ruled out and the trait factor of the individual can be tested.

Benzodiazepine being taken by the patient was noted and the patient was advised not to take the nighttime dosage before cognitive assessment. The patients who did not turn up after 2 months were contacted through telephone and asked to come for the cognitive assessment. They were traced until 6 months from the point of intake in the study.

Patients were categorized into early-onset (onset before the age of 25 years) and late-onset (onset after the age of 25 years) alcohol dependence. Externalizing psychopathology and cognitive functions were compared between the two groups applying appropriate statistical tests using computerized statistical software (SPSS for Windows, Version 16.0. Chicago, SPSS Inc.).

RESULTS

A total of 112 patients were screened to enroll 91 patients in the study. The most common reason for exclusion was years of formal education for <8 years (n = 8). The minimum requirement of education was needed for taking part in the computerized cognitive tests which have also been done by another Indian study.[16] The history of subnormal intelligence (n = 1) was excluded for the same reason. Other reasons of exclusion were patients with serious comorbid medical illness – emphysema (n = 1), alcoholic liver disease with ascites with respiratory distress (n = 1), extensively drug-resistant tuberculosis (n = 1) who needed priority medical management, age of the patient >60 years (n = 2), and refusal to give consent (n = 1). A total of 34 (37.4%) patients were dropped out as they did not turn up for the second phase of the study. Out of them, 24 discontinued treatment and were not ready to follow up and 10 could not be reached in the contact number provided during first phase of the study even after repeated attempts in the next 2–6 months from the point of intake.

A total number of 57 patients were finally enrolled in the study, out of which 26 were early onset and 31 were of late onset. All the patients were of male gender, majority of them were from the age group 26 to 35 years, Hindu religion, were graduates, married, belonging from urban population, and nuclear family. Relevant sociodemographic and clinical variables are compared in Table 1.

Table 1

Comparison of relevant sociodemographic and clinical variables between the two groups

Demographic profile	Early-onset group (n=26), n (%)	Late-onset group (n=31), n (%)	χ2/t	df	P
Age (years)
18-25	7 (26.9)	0	18.588	3	<0.001
26-35	13 (50.0)	8 (25.8)
36-45	5 (19.2)	14 (45.2)
46-60	1 (3.8)	9 (29.0)
Education
Middle-school (8th passed)	3 (11.5)	3 (9.7)	1.286	4	0.864
High-school (10th passed)	2 (7.7)	4 (12.9)
Intermediate (12th passed)	10 (38.5)	10 (32.3)
Graduate	10 (38.5)	11 (35.5)
Postgraduate and above	1 (3.8)	3 (9.7)
Mean duration of illness (years)	9.69±6.41	7.29±2.79	1.886	55	0.065
Mean dose of chlordiazepoxide equivalent (mg/day)	22±8.6	22±7.9	0.006	55	0.996
Number of patients having past history of complicated withdrawal (delirium tremens/seizures)	8	9	0.020	1	0.886

The externalizing psychopathology score was compared between the two groups in Table 2. It was observed that the patients in the early-onset group had a higher score in all the domains including the total externalizing psychopathology score, with all the differences being statistically significant.

Table 2

Comparison of externalizing psychopathology scores between the two groups

Externalizing psychopathology domain	Early-onset group (n=26)	Late-onset group (n=31)	P
Conduct disorder	4.00±3.80	1.26±0.89	<0.001
Dissocial personality disorder	3.73±3.00	1.06±0.93	<0.001
ADHD-inattention	1.65±0.98	1.00±0.73	0.012
ADHD-hyperactivity	1.85±1.01	1.19±0.87	0.009
ODD any 6 months	1.73±0.72	0.64±0.55	<0.001
ODD worst 6 months	1.62±0.75	0.52±0.51	<0.001
Total externalizing psychopathology score	14.58±7.07	5.68±1.96	<0.001

ADHD – Attention-deficit hyperactivity disorder; ODD – Oppositional defiant disorder

The performance in WCST was compared between the two groups in Table 3. It is seen that the performance of the early-onset group was poorer in all the parameters which also reached statistical significance (except in trials to complete the first category and failure to maintain set). In the “failure to maintain set” the early-onset group had a higher score in comparison to late-onset group. The comparison of mean scores of CPT showed that the performance of the early-onset group was poorer in all the parameters and reached statistical significance in omission errors, as shown in Table 4.

Table 3

Comparison of mean scores of Wisconsin Card Sorting Test between the two groups

WCST parameters	Early-onset group (n=26)	Late-onset group (n=31)	Mann-Whitney U-test	Z	P
Total correct	78.38±12.10	85.94±11.83	273.0	2.098	0.036
Total errors	41.77±18.08	28.29±14.10	235.0	2.728	0.006
Perseverative responses	22.50±11.52	14.16±7.35	217.0	3.020	0.003
Perseverative errors	21.38±9.83	13.77±6.97	217.0	3.020	0.003
Nonperseverative errors	20.38±10.12	14.52±7.54	260.0	2.345	0.019
Conceptual level responses	64.04±21.68	75.68±16.36	296.0	1.728	0.084
Categories completed	3.31±2.22	4.42±1.93	279.5	2.088	0.037
Trials to complete first category	30.27±26.77	23.45±18.89	362.5	0.663	0.508
Failure to maintain set	3.88±1.68	2.87±1.75	296.5	1.744	0.081

WCST – Wisconsin Card Sorting Test

Table 4

Comparison of mean scores of continuous performance test between the two groups

CPT parameters	Early-onset group (n=26)	Late-onset group (n=31)	Mann-Whitney U-test	Z	P
Omissions	25.35±24.64	10.61±8.89	235.0	2.697	0.007
Commissions	16±6.76	13.52±6.13	317.0	1.387	0.165
Hit rate (overall hit reaction time)	450.28±128.4	486.23±119.62	316.0	1.397	0.163

CPT – Continuous performance test

DISCUSSION

As this was a time-bound study of 9-month duration, a cross-sectional study design was followed. In this study, it was hypothesized that early-onset alcohol dependence has higher externalizing psychopathology and poorer cognitive performance. It has been shown in various studies that age-related cognitive decline occurs at ages 60 and above.[1718] Hene, patients who were older than age 60 years were excluded to remove any confounding factor in cognitive performance due to advanced age. Some other reasons for exclusion were being not accompanied by relatives who can give a reliable history (n = 5) which was needed to corroborate with the details provided by the patient in the lifetime sections of the interview. Previous studies in India regarding the age of onset have also taken such criteria to avoid any recall bias or falsification of the statement of the patient[5] and have excluded patients based on not having relatives who could give details of substance use.[6]

In our study, the dropout rate after the initial assessment was 37.36% (i.e. 34/91). Previous studies show that the dropout rate from the treatment of AUD ranges from 17.4% to 74% in the inpatient population, whereas in the outpatient population, it even reaches as high as 70%.[1920] Various benzodiazepines prescribed in the study population were chlordiazepoxide, oxazepam, and lorazepam. They were converted to chlordiazepoxide equivalent dosage as per the Maudsley prescriber's guidelines[21] and on comparing the two groups, no significant difference was found. It is to be noted that the patients were instructed not to take their last night and morning dosage of benzodiazepines when they were called for the cognitive tests, as benzodiazepine may adversely affect the cognitive performance.

The externalizing psychopathology scores were compared between the two groups as obtained from SSAGA- IV. It was seen that in all the 6 domains and the total score the early-onset group had a higher mean score with the difference being statistically significant in all domains. It is known that the early-onset group of alcohol dependence has higher externalizing psychopathology as compared to late onset which has been shown in various studies.[112223242526] The matter of interest lies in the fact about the underlying association. The finding may point out that the lower age of first alcohol intake may disrupt the normal, social, and intellectual framework and may give rise to behavioral problems. On the other hand, there remains a concept that both the manifestations of early alcohol use and externalizing psychopathology may be due to the same underlying mechanism.[26] The step that is proximal in the trajectory remains the point of interest. Even a neurobiological explanation for the same exists. The study shows that Hydroxytryptamine) “S” promoter polymorphism is associated with a higher risk for early-onset alcoholism associated with an antisocial personality disorder, impulsive, and habitually violent behavior.[27]

Comparing the mean scores of the WCST between the early-onset and late-onset group showed a poorer performance by the early-onset group in all the parameters. WCST is a test of the executive function used to study “abstraction ability” and “the ability to shift cognitive strategies.” It is a measure of cognitive flexibility and perseverative responding. This task measures concept generation, cognitive flexibility, and the lack thereof: perseveration. It requires an individual to sort a long series of cards with designs according to various principles that can only be gleaned by incorporating information learned while the performance of the test. This measure then assesses the individual's ability to generate and test various solutions, to flexibly respond to changing circumstances, as well as the ability to integrate new information with experience to respond as efficiently as possible. It is a sensitive test for executive function, especially for assessing the ability for self-correction which is defective in alcohol use population. The behavioral symptoms of decreased voluntary motor behavior, decreased energy, tendency to engage in repetitive behavior, abnormalities of emotion particularly apathy, indifference, and shallowness can be detected in WCST which forms a part of an alcoholic makeup. It is seen that the deficits in these demanding tasks can be detected only through neuropsychological assessments where alcoholics perform poorly than their nonalcoholic counterparts.[28] This may also be due to the cumulative effect of genetic problems in terms of neuropsychological deficits. The existence of a cognitive endophenotype in alcohol-dependent patients is also hypothesized. The increase in perseverative errors, nonperseverative errors, and conceptual level responses in the early-onset group can be explained through the following mechanism. The cognitive abilities of visual-spatial memory, deductive reasoning, cognitive flexibility, and problem solving which are needed for performing well in WCST are controlled by specific areas of the brain circuitry.[28] Specific prefrontal areas are involved in different stages of task performance. The mid dorsolateral prefrontal cortex (area 9/46) is involved in negative or positive feedback the point where current information is related to earlier events stored in working memory. A cortical basal ganglia loop involving the mid ventrolateral prefrontal cortex (area 47/12), the caudate nucleus, and mediodorsal thalamus has increased activity during receiving negative feedback, which is a signal for a mental shift to a new response set. The posterior prefrontal cortex is associated with specific actions to stimuli. The putamen has an increased activity after matching negative feedback showing a greater involvement during novel than routine actions.[29] It has been seen that individuals with ADHD or with problems in inattention have difficulties in several domains like problem-solving, planning, orienting, cognitive flexibility, sustained attention, response inhibition, and working memory. These constructs are also found to have a similar neurobiological involvement.[30] Thus, the early-onset group of our study who has a higher ADHD inactivity score in externalizing psychopathology scores could have performed poorly in the WCST. On the other hand, impairments in frontal limbic networks are responsible for hyperactivity symptoms. This can also modify the test performances as the early-onset group of our population had higher ADHD-hyperactivity scores. Studies have shown that alcohol-dependent patients who have a positive family history have poorer performance in WCST[31] in all parameters. There has been also a higher family history of alcohol use in the early-onset group suggesting an explanation for the worse performance in this group. These impairments may be due to the combined effects of genetic liability and the direct toxicity of alcohol on the frontal lobes. The direct toxicity can be due to the duration of illness and as found in our study the early-onset group had more duration of illness when compared to the late-onset group. It is also postulated that impulsivity and executive function impairment also has a role in alcohol dependence and their early onset.[31]

The CPT measures a person's sustained attention, selective attention, and impulsivity. The omission errors (resulting from the failure to respond to target letters) serve as a measure of sustained attention, whereas commission errors (made when responses are given to nontargets) serve as a measure of selective attention. Hit Reaction Time (Hit RT) - overall is the average speed of correct responses for the entire test. This gives a measure of impulsivity. It has been seen that ADHD patients reasonably poor in CPT. It tests both the ability to perform a repetitive task and to inhibit prepotent responses. It is seen that patients with alcohol use also have impaired CPT performance.[32] Another study says that the performance of CPT is worse in type 2 alcoholics.[33] As have been seen in studies that early-onset alcohol use may have certain attentional impairments which can worsen the performance further.[34] Our study also shows a poorer performance in CPT by the early-onset group with significantly poorer performance in omission errors (P = 0.007). Thus, the presence of intentional construct with other deficits due to toxic effects of alcohol all together combines in worse performance in the early-onset group. The externalizing psychopathology as depicted in the externalizing scores of SSAGA IV and performance in cognitive tests WCST and CPT all show a different pattern of findings in the early-onset group when compared to the late-onset group. The externalizing psychopathology score is higher in all domains (CD, antisocial personality, ADHD hyperactivity, ADHD inattention, ODD any 6 months, ODD worst 6 months, and total externalizing score) of the early-onset group. This can have an impact on all the domains of alcohol-related problems.

Identification of the subjects who are at higher risk can help early intervention and possible approaches to mitigate as early as possible. Hence, finding out any of the deficits like high externalizing behaviors in childhood or which may even progress to adulthood or deficits in any neurocognitive aspect (specifically tests of executive function) should alert any clinician. A thorough assessment of all other domains should be done in all such patients.

It has been seen that AUD contributes to morbidity and disability in a huge proportion irrespective of the age of onset. Taking alcohol and continuing using it at any level of problematic use is harmful to the individual and society as a whole. The classification of AUD based on the age of onset has been hypothesized and put on research from long back. If a simple entity like “age of onset” can be applied to the clinical population to stratify the risk group, then it can be highly useful. Our study shows valid evidence that “early-onset versus late-onset” concept exists and the early onset falls in the higher risk strata. Identification of patients with “early-onset” in a proximal phase of the trajectory can serve as a helpful measure in the prevention of further problems in the individual. Even from a perspective of public health, a systematic study of the effects of age at first alcohol use is essentially important. Prevention programs as delaying the first use of alcohol can be an important step. Wholesome management in this viewpoint can help reduce individual suffering by mitigating social, economic, and health-related costs.

The limitations of the study were that the sample was taken from a clinical setting rather than a community setting. The basis of determination of the presence of ADHD/CD/ODD was essentially and necessarily retrospective. Hence, factors such as recall bias, selective forgetting, and retrospective falsification cannot be ruled out. To minimize the effect of these factors, information put forth by the individuals was corroborated by the available informants whenever feasible. The investigator was not blind to the outcomes. Hence, the possibility of ascertainment bias could not be ruled out. To overcome some of the limitations, the onset of alcohol dependence was determined later after interpreting SSAGA and that was done after the individuals were interviewed for the later sections of SSAGA which included the presence of externalizing disorders. Hence, the interviewer was not aware of their onset of alcohol dependence status during the time of assessment. This must have reduced the interviewer's bias. Similarly, the ones eventually analyzed were those patients who remain abstinent and came to follow-up. This group would possibly vary from the study population analyzed initially. There are basically treatment completers and are compliant patients, who are likely to have better motivation, cognitive functions, and psychosocial support than the whole group. It can be a limitation in our study; however, it has been observed, both in the early-onset and late-onset groups.

CONCLUSION

Patients with early-onset alcohol dependence have higher externalizing psychopathology and more cognitive deficits in comparison to those with late-onset alcohol dependence. Further, research in a larger sample will help in establishing the relationship between externalizing psychopathology and cognitive performance in different subgroups of alcohol dependence.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.