The impact of recreational cannabis use on neuropsychological function in epilepsy
University College Hospital, London, United Kingdom
UCL Queen Square Institute of Neurology, Department of Clinical and Experimental Epilepsy, United Kingdom
⁎Corresponding author at: ESRC Epilepsy Society, Chesham Lane, Chalfont St Peter, Buckinghamshire, SL9 0RJ, United Kingdom. s.baxendale@ucl.ac.ukHighlights
- •Recreational cannabis use is reported more commonly in males than females.
- •Cannabis use is associated with lower baseline intellectual reserves.
- •Cannabis use in people with epilepsy amplifies deficits in new learning.
- •Enhanced susceptibility of recall to distraction is also evident in this group.
Abstract
Cannabis use is associated with neuropsychological impairments in the general population, but little is known about the impact on cognitive function in people with epilepsy who are already at increased risk of difficulties due to the essential comorbidities of the disease. We compared the performance of 42 people with epilepsy (PWE) who reported regular cannabis use with 254 age matched, non-cannabis-using PWE. Patients completed tests of intellectual reserve, memory, language and processing speed. Approximately one in 17 patients (5.9 %) reported current cannabis use. Cannabis use was not associated with epilepsy type. Males were 1.8 times more likely to report cannabis use compared to females. Cannabis use was associated with lower intellectual reserve (Reading IQ: t = 2.8, p < 0.01, Cohen’s d = 0.49), reduced encoding of new information (List Learning: t = 3.3, p < 0.001, Cohen’s d = 0.56) and enhanced susceptibility to distraction on a subsequent recall task (t = 3.07, p < 0.01, Cohen’s d = 0.51. In regression models cannabis use was significantly associated with impairments in learning and recall after controlling for elevated levels of anxiety and depression. Our data indicates that recreational cannabis use in people with epilepsy amplifies deficits in new learning and enhances susceptibility to distraction in the retention of newly learnt material. Recreational cannabis use should be considered when interpreting the significance of these cognitive impairments when they are recorded in a clinical assessment.
Introduction
Cannabis is the most widely used recreational drug in the world and tops the table of ‘the most used drug’ in countries across the five continents [1]. However, whilst widespread, only eight countries have legalised its use for recreational purposes (Canada, Georgia, Luxembourg, Malta, Mexico, South Africa, Thailand, and Uruguay). Recreational use has also been legalised in some states in the USA. Other legislative approaches to the drug have been to decriminalise possession and users are not actively pursued by criminal law enforcement. Whilst it is still illegal to possess and use cannabis, civil penalties rather than criminal convictions will apply. This approach is more common in Europe, with Germany, Italy, Netherlands, Spain, Portugal and Switzerland adopting this approach.
Cannabis is classified as a class B drug in the UK. Penalties for possession include up to 5 years in prison, an unlimited fine or both [2], Nevertheless, cannabis is the most widely available and commonly used recreational drug in the UK with 7.6 % of adults reporting that they regularly or sometimes use the drug [3].
The possible anticonvulsant properties of cannabis have long been recognised [4]. Whilst some cannabis-based treatments were legalised in the UK in November 2018, very few NHS centres prescribe medicinal cannabis due to the limited and low-quality evidence of its efficacy to treat seizures in adults [5], [6] Thus, some people with epilepsy report using cannabis for ‘self-medicating’ purposes, in addition to recreational use[7]. The prevalence of cannabis use in adult epilepsy populations varies according to country, ranging from 6 % in Germany to 75 % in Canada [8], [9]. See Table 1.[8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19].Study Country Prevalence Rate N Participants Additional Findings Feeney (1976) USA (New Mexico) <1% for adults over the age of 30
29 % for adults less than 30 years old98 Most cannabis users reported no effect on seizures. Gross et al. (2004) Canada 21 % were active cannabis users: 13 % frequent users and 8 % heavy users 136 48 % had used cannabis in their lifetime.
Most active users reported beneficial effects on seizures.
In the regression analysis seizure frequency, longer disease duration and other illicit drug use were predictors of frequent cannabis use.Saha et al. (2006) South Africa 24 % 101 Hamerle et al. (2014) Germany 20 % had used cannabis since epilepsy diagnosis: 4 % were frequent active users 310 Current cannabis use was predicted by younger age and being male. Massot-Tarrús & McLachlan (2016) Canada 57 % 190 84 % of cannabis users perceived improvement in seizures.
85 % reported improvement in stress.
77 % reported improvement in sleep.
32 % of cannabis users perceived improvement in memory and concentration.
51 % report unchanged cognition.
17 % report worsened memory and concentration.Suraev et al. (2017) Australia 15 % 587 Cannabis use was predicted by the number of past AEDs tried, those who had a neurological condition in addition to epilepsy, the presence of pain, and epilepsy due to a structural brain abnormality.
Most cannabis products were from illegal supplies with no awareness of its formal composition.Moores et al. (2018) Canada 30 % 140 25 % of participants had consumed cannabis in the past.
Cannabis use was associated with increased seizure frequency.
Side effects were common, the most frequent being mood.Taalab et al. (2019) Egypt 38 % 440 Most cannabis users reported that cannabis had a positive impact on their epilepsy condition. von Wrede et al. (2019) Germany 6 % 275 28 % reported past cannabis use. Wahby et al. (2019) Canada 21 % 337 Cannabis use independently associated with poor psychosocial health including depression, lower quality of life, worse epilepsy-related disability, and lower satisfaction with antiepileptic medication.
Cannabis use partially mediated the negative effect of a psychiatric history on psychosocial outcomes.Jackson-Tarlton et al. (2020) Canada 19 % 43 Daily cannabis use was common, but there was no significant difference in use depending on diagnosis (newly diagnosed epilepsy, new-onset epilepsy, first unprovoked seizure). Puteikis & Mameniškienė (2020) Lithuania 16 % 250 Cannabis use was positively predicted by the belief that cannabis is safer because of its natural origin, and the premise of understanding its legal status. These patients were also more likely to consult internet sources and believe that cannabis is an effective epilepsy treatment option with no side effects. Esmonde-White et al. (2023) Canada 75 %: 70 % use cannabis at least once daily 395 221/395 participants had used cannabis within the past year
Main reported side effects of cannabis use included impaired thinking (17 %), anxiety (16 %) and altered hunger (15 %). 60 % found cannabis was ‘somewhat’ to ‘very’ effective in reduce seizure frequency.
Cannabis users main concerns about use included financial strain (37 %), lack of recommendation from a doctor (30 %) and a lack of information around cannabis use (19 %).
These widely disparate percentages are likely to reflect a number of factors including the legal status of the drug in the country, and associated ease of availability. Methodological differences in recording cannabis use will also impact reported rates (for example some studies ask about regular use, whilst others ask about use within the past year). There are also likely to be cultural differences in the rate of reporting what is an illegal activity for many, in a clinical or research setting.
A recent review by Li et al. (2023) [20] presents a comprehensive overview of ‘non-medical cannabis’ use in people with epilepsy. In their review of the literature, the median number of adults and children with epilepsy using cannabis was 24.5 %. Factors associated with non-medical cannabis use included male sex, younger adult age and lower education status, with cannabis use contributing to stigma and higher levels of depression. The reviewers suggest the evidence-base is sparse and heterogeneous with mostly small cross-sectional studies, and there are significant gaps in the literature including determining the neuropsychological effects of recreational cannabis use in people with epilepsy.
In non-patient populations, chronic cannabis use can lead to cognitive impairment including detrimental effects on learning, memory, executive function, global cognition, attention, and decision-making [21], [22], [23], [24], [25], [26], [27]. The most commonly reported cognitive impact is in the memory domain with encoding, storage, manipulation, and retrieval of verbal information affected [28], [29], [30]. Some studies report neutral or mildly beneficial effects of medicinal cannabidiol on neuropsychological function in people with epilepsy [31], [32], [33] which may be secondary to improved seizure control. These inconsistent findings may reflect, at least in some part, the significant methodological challenges of studies in this field. In non-experimental settings it is difficult to control for the heterogeneity of recreational cannabis use with respect to duration and frequency of use, potency, age of onset and periods of abstinence [25], [30], [34], [35], [36].
THC is the psychoactive component of cannabis acting through the endocannabinoid system which regulates mood, learning, and memory [37] and has a high content within current strains of cannabis in the UK [38]. Strains with higher Δ9-tetrahydrocannabinol (THC) content are thought to result in greater cognitive impairment and mood disturbance [39].
Little is known about the additive impact of recreational cannabis use on neuropsychological function in adults with epilepsy, with this population already at-risk of memory and cognitive difficulties due to the essential comorbidities of the condition. The current study aimed to examine the impact of cannabis use on cognitive function in this group.
Methods
The records of a consecutive series of 800 patients who underwent a neuropsychological assessment in the epilepsy service at the University College London Hospital between 2019 and 2022 were examined for references to recreational cannabis use. All patients who attend for a neuropsychological assessment in our department are routinely asked about recreational drug use in their clinical interview, prior to their formal assessment with standardised tests. Information regarding recreational drug use is also recorded in the medical and neuropsychiatric assessments they undergo when assessed by the multidisciplinary team. We identified seventy patients who admitted using cannabis for recreational purposes, either in the past or at the time of the assessment, comprising 8.75 % of our referrals. Forty-seven of these patients reported current use of cannabis.
Three of the patients who reported current cannabis use had an exclusive diagnosis of non-epileptic attack disorder and were excluded from the analyses of cognitive performance. Two patients had undergone surgery prior to the assessment and were also excluded from these analyses to avoid the confound of the impact of surgery on cognitive performance. Following these exclusions, our search strategy identified 42 patients (27 males, 15 females) with epilepsy who reported current recreational cannabis use at the time of their neuropsychological assessment.
Comparator group
The neuropsychological test scores of the cannabis group were compared to those of a consecutive series of 254 (144 males, 110 females) age-matched, non cannabis using people with epilepsy who underwent an assessment over the same time period. Since none of the cannabis group had an intellectual disability, patients with an intellectual disability were excluded from the comparator group. Intellectual disability was established as part of the clinical neuropsychological assessment which consisted of medical and educational review and the completion of formal tests of cognitive function. As with the cannabis group, patients who were referred for a postoperative assessment were excluded.
Neuropsychological measures
The neuropsychological assessment included a standardised measure of pre-morbid ability, the National Adult Reading Test [40] and tests of intellectual, memory, language and executive function, in addition to questionnaire data. The scaled scores from the Digit Span and Coding tests from the Wechsler Adult Intelligence Scale (WAIS-IV)[41] were used as measures of working memory and processing speed respectively. Measures of memory included the List 1–5 score from the BMIPB II [42] list learning task (a measure of learning over 5 trials) and the List A6 score (a measure of retention following distraction). The design learning score from the BMIPB was used as our measure of visual memory. The total number of items correctly named on the Graded Naming Test was used as a measure of language function. The number of words generated in 60 s beginning with ‘S’ and number of animals generated were our measures of phonetic and sematic fluency. Self-report measures included the Hospital Anxiety and Depression Scale (HADS) [43] and the Subjective Memory Questionnaire (SMQ). All of these measures have been described in detail previously[44].
The tests that comprise a clinical neuropsychological assessment are determined on an individual basis and are selected in response to the specific referral received. Not all patients undertake all tasks. This is reflected in the numbers in each analyses.
Ethical Approval
All data was fully anonymised prior to the analyses in order to conform to ethical approvals granted for the analysis of routinely collected clinical data in an audit of recreational cannabis use in patients referred for a neuropsychological assessment. (Hospital Board Approval: 20–202223-SE).
Data availability statement
The clinical data analysed in the current study are not publicly available due to patient privacy and restricted access, but further information about the database is available from the corresponding author on reasonable request.
Statistical analyses
All analyses were conducted using SPSS v27. Differences between the cannabis group (n = 42) and the comparator group on the neuropsychological measures were explored with independent t tests. Linear regression models were used to examine the relative contributions of cannabis status and mood to measures of cognitive function, where preliminary analyses indicated significant group differences in function.
Results
Demographic & clinical characteristics of cannabis users
8.75 % of the consecutive series of 800 patients who attended our department for a neuropsychological assessment reported past or present recreational cannabis use. Approximately one in 17 patients (5.9 %) reported current cannabis use at the time of the assessment (n = 47). In the consecutive series of 800 patients, cannabis use (current/past/never) was not associated with epilepsy type (Focal, Generalised, Unclassified, Non Epileptic Attack Disorder, Dual Diagnosis) Chi square = 14.9, df 8, p > 0.05.
Typically more females are referred for an assessment in our department than males (56.3 %). However, in our comparator group, which was age-matched to the cannabis group and which excluded surgical patients, males represented 56 % of the group. In the cannabis group, males represented 64 % of the group (chi square 0.8, df 1, p > 0.05).
Males were approximately 1.8 times more likely to use cannabis compared to females although this just failed to reach statistical significance (Males n = 27; Females n = 15, χ2 = 3.4, df 1, p = 0.06). Patients in our series who reported current cannabis use were also significantly younger than past and non-cannabis users (f (2,521) = 8.4, p < 0.001) with a small-moderate effect size (η2 = 03).
Neuropsychological test scores
The clinical and demographic characteristics of the cannabis group and the age matched comparator group are presented in Table 2. Cannabis use was associated with lower intellectual reserve (NART IQ) reduced verbal learning (List Learning) and enhanced susceptibility to distraction on a subsequent recall task (List A6). The cannabis group also reported significantly higher levels of anxiety and depression on the HADS. No significant differences between the cannabis group and the comparator group were evident on tests of working memory (Digit Span), processing speed (Coding), immediate or delayed prose recall, visual learning, naming or verbal fluency scores. See Table 3.Epilepsy Group (n = 42) Control Group (n = 254) Age 30.38 (9.2) 30.95 (9.7) Sex 27 males, 15 females 144 males, 110 females Epilepsy Type Focal n = 31
Generalised n = 6
Unclassified n = 2
Dual diagnosis n = 3Focal n = 160
Generalised n = 36
Unclassified n = 45
Dual diagnosis n = 13Variable Group N* Mean
(std Dev)Z score** Levene's F Levene's Sig. t-value df Sig. Cohen's d Reading IQ Non cannabis 189 99.60
(8.97)−0.02 1.10 0.30 2.82 226 0.003 0.49 Cannabis 39 94.97
(10.95)−0.33 Digit Span Non cannabis 215 7.83 (3.8) −1.44 0.02 0.88 −0.75 251 0.23 −0.13 Cannabis 38 8.24
(2.98)−1.17 Processing speed (Coding) Non cannabis 220 7.68
(3.1)−1.54 3.92 0.05 1.29 253 0.10 0.23 Cannabis 35 6.91
(2.93)−2.06 Verbal Learning (list) Non cannabis 245 45.32
(11.84)−0.82 0.48 0.49 3.36 285 <0.001 0.56 Cannabis 42 38.76
(10.77)−1.45 Verbal recall (after distraction) Non cannabis 245 9.18
(3.62)−1.7 2.86 0.09 3.07 285 0.001 0.51 Cannabis 42 7.36
(3.04)−2.5 Story (immediate recall) Non cannabis 183 18.79
(9.46)−1.18 0.44 0.51 1.28 203 0.10 0.29 Cannabis 22 16.09
(8.17)−1.46 Story (delayed recall) Non cannabis 181 16.67
(10.78)−1.19 1.17 0.28 1.20 201 0.12 0.27 Cannabis 22 13.82
(7.75)−1.49 Visual Learning Non cannabis 227 36.39
(10.77)−0.16 5.69 0.02 1.17 265 0.12 0.20 Cannabis 40 34.30
(8.27)−0.36 Graded Naming Non cannabis 209 14.06
(5.03)−1.54 0.76 0.38 −0.74 234 0.23 −0.15 Cannabis 27 14.81
(4.77)−1.36 Fluency (semantic) Non cannabis 228 18.74
(6.43)−0.25 0.02 0.89 0.98 266 0.16 0.17 Cannabis 40 17.65
(6.76)−0.51 Fluency (phonetic) Non cannabis 225 12.16
(5.58)−0.83 1.70 0.19 −0.84 263 0.20 −0.14 Cannabis 40 12.98
(6.25)−0.77 HADS Anxiety Non cannabis 202 9.60
(4.55)– 3.74 0.05 −2.62 238 0.01 −0.46 Cannabis 38 11.66
(3.74)– HADS Depression Non cannabis 202 5.87
(4.19)– 0.06 0.80 −3.32 238 <0.001 −0.59 Cannabis 38 8.34
(4.31)– Subjective Memory Score Non cannabis 195 61.56
(19.96)– 0.24 0.63 −1.46 227 0.07 −0.27 Cannabis 40 56.15
21.75)–
Since elevated levels of anxiety and low mood can contribute to poor scores on neuropsychological tests we examined the relative contributions of cannabis use and the HADS Anxiety and HADS depression scores to the List learning and List Recall scores via linear regression.
List learning: Regression analyses
A univariate linear regression analysis was conducted to examine the relationship between verbal learning as the dependent variable (List Learning A1-5) and the independent variables: “HADS anxiety score,” “HADS depression Score”, “premorbid level of function” (NART IQ) and “cannabis use group” (Yes/No).
The results of the analysis are presented in Table 4. The corrected model, which includes the independent variables “HADS Anxiety,” “HADS Depression,” “NART IQ,” and “Cannabis Group,” was found to be significant, F(4, 198) = 7.992, p < 0.001. This indicates that the model as a whole had a significant effect on the dependent variable “List 1–5.” Specifically, “Nart IQ” (F = 21.455, p < 0.001) and “Cannabis Group)” (F = 4.326, p = 0.039) made significant contributions to the model, while “HADS Anxiety” (F = 0.049, p = 0.824) and “HADS Depression” (F = 0.340, p = 0.560) were not significant predictors. The overall model explained a small, but significant proportion of the variance in “List 1-5″ (R2 = 0.139, Adjusted R2 = 0.122).Type III Sum of Squares df Mean Square F Sig. Corrected Model 3301.455a 4 825.364 7.992 <0.001 Intercept 70.312 1 70.312 0.681 0.41 HADS Anxiety 5.111 1 5.111 0.049 0.824 HADS Depression 35.136 1 35.136 0.34 0.56 NART IQ 2215.628 1 2215.628 21.455 <0.001 Cannabis Group 446.713 1 446.713 4.326 0.039 Error 20447.235 198 103.269 Total 443,691 203 Corrected Total 23748.69 202
Recall following distraction: Regression analyses
A second univariate linear regression analysis was performed to investigate the relationship between the dependent variable “List A6″ and the independent variables ”HADS Anxiety,“ ”HADS Depression,“ ”NART IQ,“ and cannabis use group (Yes/No). The results of the analysis are presented in Table 5. The corrected model, including the independent variables ”HADS Anxiety,“ ”HADS Depression,“ ”NART IQ,“ and ”Cannabis Group,“ showed statistical significance, F(4, 198) = 5.667, p < 0.001. This suggests that the model as a whole had a significant impact on the dependent variable ”List A6.“ As with the predictors of verbal learning (List 1–5) ”NART IQ“ (F = 11.934, p < 0.001) and ”Cannabis group“ (F = 5.316, p = 0.022) made statistically significant contributions to the model, whereas self-reported mood scores on the HADS did not: ”HADS Anxiety“ (F = 0.012, p = 0.912); ”HADS Depression“ (F = 0.294, p = 0.588). Whilst the overall model accounted for a statistically significant portion of the variance in ”List A6″ (R2 = 0.103, Adjusted R2 = 0.085), the portion was small.Type III Sum of Squares df Mean Square F Sig. Corrected Model 232.475a 4 58.119 5.667 <0.001 Intercept 0.093 1 0.093 0.009 0.924 HADS Anxiety 0.125 1 0.125 0.012 0.912 HADS Depression 3.014 1 3.014 0.294 0.588 NART IQ 122.393 1 122.393 11.934 <0.001 Cannabis Group 54.52 1 54.52 5.316 0.022 Error 2030.589 198 10.256 Total 19,841 203 Corrected Total 2263.064 202
Discussion
In our tertiary referral centre, 8.75 % of patients referred for a neuropsychological assessment reported past or present cannabis use, with 5.9 % reporting current use. Cannabis use was more commonly reported by males than females and the cannabis users were typically younger than the wider cohort. The cannabis group displayed lower intellectual reserve, reduced verbal learning abilities, and increased susceptibility to distraction on a verbal recall task compared to the matched comparator group, in a pattern consistent with that reported in a recent comprehensive review of cannabis use in this population [20]. The patients in the cannabis group reported higher levels of anxiety and depression on the HADS. Linear regression analyses revealed that whilst premorbid level of intellectual function was a significant predictor of performance, cannabis use was also independently associated with poorer verbal learning and recall following distraction, even after controlling for elevated levels of anxiety and depression.
Our results are consistent with previous studies that have shown associations between cannabis use and cognitive impairments in healthy individuals, particularly in the verbal learning and memory domains [25], [28], [30]. Whilst several studies have reported that cannabis use, particularly heavy and long-term use, is associated with deficits in various cognitive domains, including attention, memory, executive function, and processing speed [21], [22], [23], [24], [26], [45] the deficits recorded in our sample were primarily limited to the verbal memory and attention domains. The increased susceptibility to distraction on a recall task among cannabis users aligns with research suggesting that cannabis use can impair attentional processes and is associated with difficulties in maintaining attention and inhibiting distractions [22], [28]. However it is also possible that poor performance on recall following distraction was a hangover from the original, shallow encoding across the five trials of the list learning task in out sample. We did not have the final trial score disaggregated from the total list learning score in this dataset available to test this hypothesis further. Further work is underway to examine the relationship between the final trial score on the list learning task and the subsequent recall score in the people with epilepsy and we plan to include cannabis use as a variable of interest in this wider analysis.
It is interesting that it is the language dominant memory networks that appear to be primarily impacted by cannabis use in our sample. We have observed a similar pattern when we have looked at the exacerbatory impacts of anxiety and low mood on neuropsychological test performance in people with epilepsy [46]. These functions appear to be particularly vulnerable to factors that impact on the cognitive reserves of people with epilepsy, which may already be compromised due to organic factors associated with seizures and the impacts of anti-seizure medications. However, whilst this pattern is seen in relation to other factors that impact cognitive function, it is unclear why language dominant networks would be primarily affected. It may reflect something specific to the language dominant networks themselves or may be an artefact of our test construction. Our tests of verbal ability may tap into more discreet networks, whilst performance on tests of non-verbal ability may be more widely distributed, allowing for more pressure on the system before a deficit becomes evident. Insights from functional imaging paradigms may be able to elucidate this further.
There are many reasons why people with epilepsy experience difficulties in multiple cognitive domains, including the underlying pathology responsible for their seizures, the effects of seizures and sub-clinical EEG abnormalities, the impacts of antiseizure medications and common comorbidities associated with the condition [47], [48]. The fact that we did not find impacts of cannabis use on a wider range of cognitive functions, as would be predicted by the literature in healthy individuals may reflect the fact that these functions are already often compromised by epilepsy related variables [49]. It is also worth noting that the cognitive impairments associated with cannabis use can be subtle and may not be evident in all individuals or under all conditions [50]. Factors such as age, duration and frequency of use, potency of cannabis, and individual differences in vulnerability are likely to influence the magnitude and persistence of cognitive deficits.
The elevated levels of anxiety and depression in our cannabis group is consistent with previous research linking cannabis use to increased psychological distress [51]. Self medicating with cannabis for anxiety and low mood is often reported as a primary motivation by people who use the drug on a recreational basis [52]. Our study design does not allow for casual inferences and we can only note the increased association of cannabis use in our group in young men with limited cognitive reserve. It is well established that men are less likely to access mental health services as readily as women, with stigma and help-seeking attitudes identified as key facets contributing to males seeking less support for mental health difficulties [53], [54]. In the UK, a disproportionately low number of males compared to females are referred to mental health services and receive mental health treatments [55]. As such, a tendency for males to self-medicate with alcohol and drugs to ease emotional distress has been recognised [53], [54] which may explain the association between males and ‘self-medicating’ cannabis use within the present study. Additional barriers to accessing mental services are associated with limited cognitive reserve. In addition to cognitive factors, other barriers to accessing mental health services may include long waiting lists and inadequate provision resulting in individuals turning to alternative remedies for their difficulties.
However it is also possible that cannabis use has a bidirectional relationship with lower cognitive reserve. Whilst those with more limited cognitive reserve may be more likely to self-medicate due to the barriers they face accessing mental health services, cannabis use may also result in more limited cognitive reserves, particularly if people began using it during their adolescence, when neurodevelopment was still underway. Longitudinal studies have shown that cannabis use during adolescence and young adulthood, when the brain is still developing, may have a more pronounced negative impact on cognitive abilities [27]. Persistent cannabis use in adolescence or young adulthood may interfere with critical brain development processes resulting in global blunting of neuropsychological function, as reported by Meier et al. (2012) [27]. Due to the existing vulnerability of brain function in people with epilepsy, the additional neuropsychological burden of cannabis use during young adulthood may have negative impacts on the development of a broad range of neuropsychological functions. This could possibly account for our finding that the cannabis users within this study had significantly lower reading IQ than non-cannabis users. Further data establishing the age of onset of cannabis use will be needed to investigate these possibilities.
Limitations
There are several limitations of the study. The data relied on self-report measures of recreational cannabis use. Given that cannabis use remains illegal in the UK, patient reports are likely to underestimate true cannabis use in this sample. Moreover, the sample consisted of patients who were referred for a clinical assessment of their cognitive function. Whilst the reasons for their referral varied, all patients will have been seen due to concerns about cognition, arising from the patient, their family or the treating clinician or in the context of a presurgical assessment for medically intractable epilepsy. These patients are not representative of the majority of people with epilepsy whose seizures are well controlled on medication and who are never referred for a formal neuropsychological assessment. The pragmatic sample and observational design limit our ability to establish causal relationships or assess the long-term or cumulative effects of cannabis use. Caution should therefore be used in generalising these findings to the broader population.
Future research
Whilst the legal status of recreational cannabis currently prohibits any kind of controlled trial to investigate the impact of recreational, non-prescribed use on cognitive function in many countries, such research may be possible in jurisdictions where the use of cannabis has been legalised, although ethical considerations may prevent any study designs involving randomisation, even in these settings. In other regions where recreational cannabis remains proscribed, follow-up assessments tracking the cognitive performance of those who report cannabis use to examine potential recovery or reversibility of cognitive deficits with cessation, would provide valuable insights to share with this patient population. Whilst we did not find any association between epilepsy type and cannabis use, further investigations into the interactions between cannabis use and different anti-seizure medications and impact on cognitive function may be a useful area of research. In a larger sample, the exploration of potential dose–response relationships and the impact of different cannabis strains and modes of consumption could also further our understanding of the cognitive effects of cannabis in this patient group.
Conclusions
In summary, we found that approximately one in 17 of the patients referred to our service reported recreational cannabis use. Usage was most commonly reported by young males with elevated levels of anxiety and depression. Recreational cannabis use was associated with lower intellectual reserve, reduced verbal learning, and increased susceptibility to distraction on a recall task. These effects remained significant even after controlling for elevated levels of anxiety and depression. The findings underscore the importance of considering cannabis use as a potential factor influencing cognitive function in neuropsychological assessments, particularly in presurgical settings where deficits in verbal memory may be misattributed.
Ethical statement
This study examined patient data collected on a routine clinical basis. All data was fully anonymised prior to the analyses in order to conform to the ethical approvals granted for the analysis of this data an audit of recreational cannabis use in patients referred for a neuropsychological assessment. (Hospital Board Approval: 20–202223-SE). No individuals can be identified from the information presented in this study. The clinical data analysed in the current study are not publicly available due to patient privacy and restricted access, but further information about the database is available from the corresponding author on reasonable request.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
Data will be made available on request.