The relation between depression, sleep and cannabis use in university students

Abstract

The relationship between depression, sleep and cannabis use has been reported to have variable outcomes within the available literature thus far. The relationship between said variables has not, to date, been analysed within a student sample, which has been the goal of this questionnaire-based study. A total of 73 students (15 male; 58 female) at the University of Manchester completed all of the necessary questionnaires required to analyse the relationship between depression, sleep and cannabis use in a student sample. To test these variables, several questionnaires were employed: the Hospital Anxiety and Depression Scale (HADS) to test depression symptoms; the Pittsburgh Sleep Quality Index (PSQI) to test sleep quality; the ‘Cannabis Use Disorder Identification Test - Revised’ (CUDIT-R) to test problematic cannabis use; a ‘weekly joints’ questionnaire to test cannabis use frequency and finally a ‘cannabis as a sleep aid’ question to test whether students who used cannabis as a sleep aid had better or worse subjective sleep quality and to test whether they had more or fewer symptoms of depression when compared to non-users. The study found a strong positive correlation between problematic cannabis use (CUDIT-R) and number of joints consumed weekly, as well as a significant positive correlation between depression (HADS) and sleep quality (PSQI). It was also found that non-users had a better sleep quality and less severe symptoms of depression than student cannabis users.

Introduction

Among the variety of psychoactive substances, cannabis is an exceedingly popular substance used by many individuals in the UK. Cannabis was the single most used substance among 16 to 59 year olds in 2015/2016, with 2.1 million individuals having used cannabis and 11% of 15-year-olds having used the substance in 2014 (Health & Social Care Information Centre, 2016). Cannabis is used by many for a variety of reasons such as positive mood alteration, coping with negative emotions, to aid social activity, relaxation (Green, Kavanagh & Young, 2004) and as a ‘sleep aid’ (using cannabis as a means to enter into a sleep state) (Bonn-Miller, Babson & Vandrey, 2014). The use of cannabis as a sleep aid usually involves the ingestion of cannabis through smoking, eating or vaporising during the time period before sleep in order to ease oneself into a sleep state. Research has shown that pre-sleep cannabis use decreases the time spent in stage 3 (slow-wave) and stage 4 (REM) of the sleep cycle, subsequently decreasing overall sleep quality (Freemon, 1974). Despite pre-sleep cannabis use actually decreasing sleep quality, cannabis is often used by individuals as a sleep aid because ingestion of the substance decreases sleep onset latency (Chait, 1990; Cousens & Dimasco, 1973), that is: time taken to accomplish the transition from full wakefulness to early non-REM sleep stages (Nicholson, Turner, Stone & Robson, 2004). This sleep onset latency effect often misleads cannabis users into believing that cannabis use improves sleep quality, when research has shown that this is not the case. Stage 3 (slow-wave) and stage 4 (REM) of the sleep cycle are the most important stages of the sleep cycle, in terms of psychological development and processing. Stage 4 (REM) sleep is the stage during which dreaming occurs; this stage has been hypothesized as providing stimulation to regions of the brain involved in the consolidation of emotional information and the processing of sensory information, namely the hippocampus and the temporal-occipital areas ("Neuroscience of sleep", 2017). Because of the nature of using cannabis as a means to help one fall asleep, users may indeed be misled to believe that cannabis is beneficial to the aforementioned stages of sleep, when empirical evidence has long shown that cannabis use actually has a negative impact on sleep quality (Freemon, 1974). In parallel, research on medical cannabis users suggests that cannabis use improves perceived sleep quality (Babson, Boden & Bonn-Miller, 2013), further providing evidence of this misled belief. Interestingly, these researchers found that the largest amount of cannabis was used by individuals who had the highest levels of depression and the best perceived sleep quality (Babson, Boden & Bonn-Miller, 2013). Exploring the relationship between cannabis use, sleep quality and depression in individuals who use cannabis as a sleep aid will surely provide interesting results, especially when compared to individuals who do not use the substance as a sleep aid.

The relationship between cannabis use and depression has been reported to have variable outcomes in the literature. Research suggests that the effects that cannabis has on mental health depends on the frequency and amount of cannabis that is ingested. Cannabis has been shown to be an effective anti-depressant when used in low doses, however this effect is reversed when the substance is used in high doses (McGill University, 2007). There are at least 113 different phytocannabinoids (naturally-occurring chemical compounds) within the cannabis plant, all of which can have slightly varying effects on the brain when activated (Aizpurua-Olaizola et al., 2016). This makes investigation of the cannabis-depression relationship difficult; analysing the phytocannabinoids that are present in different strains of cannabis and isolating them to determine which phytocannabinoids are associated with specific symptoms of depression has never before been accomplished. Despite this difficulty, existing research into the cannabis-depression relationship has provided interesting results. A 2003 review of the literature found that there exists a moderate association between problematic cannabis use and depression in the general population (Degenhardt, Hall & Lynskey, 2003). The review concluded that even though there is an association, it would be unscientific to rule out the possibility that the association exists due to social and contextual issues that increase cannabis use and depression. However, if the association were to be causal, Degenhardt, Hall and Lynskey agreed that the association would only be modest in nature. A more recent 2014 systematic analysis of the existing research on this cannabis-depression relationship found that, based on 57 academic articles, there may be an increased risk for cannabis users in terms of developing depressive disorders (Lev-Ran et al., 2014). In order to adjust for the aforementioned social and contextual factors, this longitudinal, epidemiological, study including 8,759 adults should be taken into account. The overall results of this study showed that after adjusting for baseline differences between adults who use cannabis and adults who do not use cannabis, any association between cannabis use and depression was significantly reduced and mostly invalidated (Harder, Morral and Arkes, 2006).

To focus on a specific, important sample, a UK-based student sample was chosen for this study. This decision was inspired by previous research showing that students are susceptible to substance misuse and mental health issues (Deas & Brown, 2006) as well as poor sleep quality (Lopes, Milheiro & Maia, 2013). It is particularly important to investigate this relationship on a student sample, as university is a time where most individuals will experience mental health issues for the first time and ingest illicit substances for the first time. University is a time during which students often experience symptoms of depression, with 27% of students experiencing some form of mental health issue; 77% of whom experience depression-related problems (Aronin & Smith, 2016). Cannabis use, according to the Cannabis Use Disorder Identification Test - Revised (CUDIT-R) (Adamson et al., 2010), is considered to be problematic when an individual produces high scores in terms of the following aspects: frequency of cannabis used, time spent being “stoned”, inability to stop using cannabis, failure of meeting expectations, time spent recovering from use, memory or concentration problems, using cannabis in physically hazardous situations and thoughts about stopping use.

The literature regarding these variables shows that the causal direction of the relationships between depression, sleep quality and cannabis use remains unclear. Currently, there exists a large literature on the relationship between cannabis use and sleep quality, the relationship between depression and sleep quality, and the relationship between cannabis use and depression. However, there is currently insufficient research regarding the association between cannabis, sleep quality and depression with these variables analysed in the same study, with the exception of Babson, Boden and Bonn-Miller (2013), who studied this association on a largely male population of medical cannabis users. They sought to further develop research on the relationship between problematic cannabis use and symptoms of depression by examining the moderating role of self-reported sleep quality within this relationship. The results of their study indicated that individuals who use cannabis on a problematic level may also experience a high level of depression due to the individuals’ perceptions that cannabis use benefits sleep quality (Babson, Boden and Bonn-Miller (2013). I aim to explore this association between cannabis use, sleep quality and depression symptoms using a UK-based student sample, as well as exploring this association in a novel context pertaining to students who use cannabis as a sleep aid and students who do not.

The posed research question is: “what are the effects of cannabis use on sleep quality and symptoms of depression in students, especially when cannabis is used as a sleep aid?”, including the following hypotheses:

1. Students with the most problematic cannabis use have the most severe symptoms of depression
2. Students with the most problematic cannabis use have the worst sleep quality
3. Students who use cannabis as a sleep aid have the best self-reported sleep quality
4. Students who use cannabis as a sleep aid experience worse depression symptoms than non-users

Method

Design & Participants

This was a cross-sectional questionnaire study for which 250 student participants were recruited using the SONA SEPS online student credit system: a university research-based system which allows researchers to advertise studies and students to complete said studies in exchange for online SEPS credits needed for the completion of their degrees. The inclusion criteria was that participants must have consumed cannabis at least once in the last three months.

Materials

A number of questionnaires were used to obtain participant data. The Hospital Anxiety and Depression Scale (HADS) (Snaith and Zigmond, 1986) was used as a measure of depression symptoms. This scale has been widely used by doctors to determine patients’ levels of anxiety and depression. HADS depression scores are categorised as follows: normal 0–7, mild 8–10, moderate 11–14, and severe 15–21. Concerning the HADS, the depression scale has an internal consistency and a reliability coefficient (Cronbach’s alpha) of .78 (Bocéréan and Dupret, 2014). The Pittsburgh Sleep Quality Index (PSQI) (Buysse et al., 1989) was used to assess several sleep-related variables including perceived sleep quality, sleep-onset latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleep medication and daytime dysfunction. The questionnaire has been used in both research and clinical settings as a method of assessing sleep quality and diagnosing sleep disorders. Scores of 5 or less out of 21 are associated with good sleep quality and any scores above 5 are associated with poor sleep quality. The PSQI has a Cronbach's alpha score of 0.83 for its seven components (Carole Smyth, 2012). The ‘Cannabis Use Disorder Identification Test - Revised’ (CUDIT-R) is a widely accepted and psychometrically reliable tool to identify problematic cannabis use within the clinical population (Adamson et al., 2010). It was used in this study to analyse the problematic nature of cannabis use in students, within which scores of 6 points or more indicate hazardous cannabis use and scores of 12 or more indicate a possible cannabis use disorder which should be explored further with an expert (Annaheim, Rehm and Gmel, 2008). The CUDIT-R has a Cronbach's alpha score of 0.914 (Adamson et al., 2010). Further, a question determining whether students use cannabis as a sleep aid was also administered: “During the past month, how often have you used cannabis to help you sleep?”. Also, a set of questions regarding demographic information and a set of questions regarding a more general use of substances was administered.

Procedure

Participants were presented with a study information sheet and were required to complete an informed consent form which ensured that participants were aware that their data will be kept confidential, that they have the right to withdraw from the study at any given time and to ensure that they definitely wanted to take part in the study. Students then confirmed their sign-up to take part of the study using the SONA SEPS system, directing them to our external survey hosted on the website: https://survey.ls.manchester.ac.uk/, powered by a survey system called ‘SelectSurvey’. Participants were initially presented with demographics questions, namely to identify gender, age, race, living situation and year of study. Subsequently, participants were asked a set of questions regarding general substance use i.e. “In your life, which of the following substances have you ever used?”. A question regarding the frequency of cannabis use was then presented: “Please tell us how many joints / spliffs you smoke each day”, alongside a table wherein participants were able to input the number of joints/spliffs smoked for every day of the week. The CUDIT-R questionnaire was subsequently presented, which included 9 questions regarding problematic cannabis use. The next questionnaire was the Pittsburgh Sleep Quality Index (PSQI) which asked 13 questions regarding the aforementioned sleep quality variables.The questionnaire following this was the Hospital Anxiety and Depression Scale (HADS), which asked 15 questions regarding levels and symptoms of depression and anxiety. Finally, participants were thanked for completing the questionnaires and asked whether their data could be used for further research or not. A number of helplines were also listed for ethical purposes: the University of Manchester Counselling Service, Saneline (a national out-of-hours telephone helpline offering emotional support and information for people affected by mental health problems) and Addaction (working solely in the field of drug and alcohol treatment, including specialist services for young people and access to local projects).

Ethical considerations

The primary ethical consideration in this study is that cannabis is a class B illegal substance in the United Kingdom. This makes it a fundamentally difficult topic to study at an undergraduate level, as there is no way of determining the validity of an individual’s responses in relation to the amount or type of cannabis they have ingested. Also, completion of the substance use questions within the questionnaire may have led to an increased awareness of problematic use amongst students, potentially inciting worry or stress about use. To combat this, all participants were signposted to Addaction, a UK-based drug and alcohol treatment charity at the end of the survey (see Appendix 7).

Furthermore, the study did address the sensitive topic of depression. Answering the Hospital Anxiety and Depression Scale (HADS) questionnaire may have caused distress to participants, as their own mental health issues may be highlighted, potentially causing an acute sense of worry or an acute exacerbation of symptoms of depression. To combat this, all participants were signposted to the University Counselling service and Saneline (a mental health helpline) at the end of the survey (see Appendix 7).

In addition, ethical approval for this study was granted for this study through the online Ethical Review Management (ERM) system. This system is in place to manage the ethical review of research projects that involve human participants and is available to both members of staff and students art the University of Manchester. It allows for applicants to submit and track their applications as well as make it easier for the University Research Ethics Committee and various panels to conduct reviews, provide feedback and approve applications.

Data analysis

A preliminary data-related issue was addressed before any analyses were conducted. Missing data was found when coding the CUDIT-R, as the survey was falsely designed, allowing participants to skip questions. To accommodate for this, the Multiple Imputation SPSS function was used to create a revised dataset for the CUDIT-R whereby participants who missed a maximum of 2 questions were given revised scores based on all other completed question scores. This process increased CUDIT-R completion from 51 participants to 73 participants.

The initial step of data analysis was to determine whether data were normally distributed. All variables (sleep quality, depression and problematic cannabis use and cannabis use frequency) were analysed for normality using the Shapiro-Wilk test on SPSS (see Table 2). After establishing normality, the next step was assessing the key variables of the study in terms of their median scores and frequencies and comparing these scores in terms of demographic variables (see Table 1). This was done by using the ‘Frequencies’ function on SPSS for each subset of data.

Subsequently, all variables were tested for correlations using the Spearman’s Rank Correlation Coefficient statistical test, as all data was not normally distributed (see Table 3).

During this process, components 1 (subjective sleep quality) and 2 (sleep onset latency) were isolated from the PSQI questionnaire and tested independently to examine correlations with HADS depression, CUDIT-R, and cannabis frequency (weekly joints) scores.

In order to test differences in variables between students who used cannabis as a sleep aid and non-users, Wilcoxon Signed Rank Tests were employed. This SPSS function tested for significant differences in terms of overall sleep quality (total PSQI scores), subjective sleep quality (component 1 of the PSQI), sleep onset latency (component 2 of the PSQI) and depression symptoms (HADS) scores.

Because a total of 38 answers to the question: “During the past month, how often have you used cannabis to help you sleep?” were spread out across ‘rarely’, ‘sometimes’, ‘often’ and ‘always’, this question was recoded into a dichotomous variable, within which revised scores were as follows: ‘no = 0’ (including scores of ‘never’) and ‘yes = 1’ (including scores of ‘rarely’, ‘sometimes’, ‘often’ and ‘always’). Recoding this variable made statistical analysis possible in terms of establishing differences in scores between cannabis users who use cannabis as a sleep aid and non-users.

Results

Participants

A total of 73 participants completed all questionnaires fully, 15 of whom were male and 58 of whom were female. The composition of ethnicity included 59 ‘white’ participants and 14 participants of ‘other’ ethnicity, as follows: 2 ‘mixed - white and caribbean’, one ‘mixed - white and black african’, 2 ‘mixed - white and asian’, one ‘other mixed background’, one ‘asian or asian british - indian’, asian or asian british - pakistani’, 4 ‘any other asian / asian british’, one ‘black or black british - caribbean’ and one ‘black or black british - african’ participant.

Primary analyses

The following variables were explored in this study: sleep quality, depression and cannabis use. Table 1 shows initial, descriptive results regarding these variables in relation to demographic variables (gender and race).

Abbildung in dieser Leseprobe nicht enthalten

Table 1: Descriptive results regarding key variables.

*’Other’ refers to any race which is not considered ‘white’ in this sample.

Testing the data for normality was the initial process in terms of data analysis. Using SPSS, Shapiro-Wilk tests for normal distribution were employed for the following variables: the ‘HADS’ for depression, the ‘CUDIT-R’ for problematic cannabis use, a ‘Weekly joints scale’ for frequency of cannabis use and the ‘PSQI’ for sleep quality. Shapiro-Wilk test were employed to show that HADS depression scores, PSQI scores, CUDIT-R scores and ‘weekly joints’ scores were not normally distributed (see Table 2).

Abbildung in dieser Leseprobe nicht enthalten

Table 2: Shapiro-Wilk scores for tests of normal distribution

As can be seen in Table 3, there was a strong positive correlation between problematic cannabis use (CUDIT-R) and number of joints consumed weekly, rho = .743, p = .000. There was also a significant positive correlation between depression (HADS) and sleep quality (PSQI), rho = .418, p = .000. No other significant correlations between variables were found.

Abbildung in dieser Leseprobe nicht enthalten

Table 3: Spearman’s rho correlations conducted between all key variables

Hypothesis testing

1. Students with the most problematic cannabis use have the most severe symptoms of depression

In order to test Hypothesis 1, a Spearman’s rho test was conducted to determine a correlation between HADS depression scores and problematic cannabis (CUDIT-R) scores. No significant correlation was found between these two variables, rho = .102, p = .391. Furthermore, as only one participant scored above 10 on the HADS depression scale with a score of 15, no relevant tests were carried out in regards to ‘moderate’ or ‘severe’ HADS scores.

2. Students with the most problematic cannabis use have the worst sleep quality

In order to test Hypothesis 2, a Spearman’s rho test was conducted to determine a correlation between sleep quality (PSQI) scores and problematic cannabis (CUDIT-R) scores. No significant correlation was found between problematic cannabis use and overall sleep quality, rho = .171, p = .147. After recoding CUDIT-R scores: normal (<6.00) and problematic (>6.00) and PSQI scores: normal (<5.00) and poor (>5.00), Spearman’s rho analysis was repeated, rho = -.062, p = .302.

When ‘Component 1’ of the PSQI, based on the question: “During the past month, how would you rate your sleep quality overall?”, was isolated and correlated with CUDIT-R scores, Spearman’s rho found a moderate correlation between subjective sleep quality and problematic cannabis use, which just failed to be significant, rho = .316, p = .057.

When ‘Component 2’ of the PSQI, based on the question, “How long (in minutes) has it taken you to fall asleep each night?”, was isolated and correlated with CUDIT-R scores, no significant correlation was found between sleep onset latency and problematic cannabis use, rho = -.033, p = .390.

3. Students who use cannabis as a sleep aid have the best self-reported sleep quality

A total of 38 participants stated that they use or have used cannabis as a sleep aid. Contrary to this hypothesis, cannabis users who used cannabis as a sleep aid had a worse sleep quality than non-users. Participants who used cannabis as a sleep aid had a median PSQI score of 7.00 (range = 13.00), while non-users had a median PSQI score of 6.00 (range = 16.00). A Wilcoxon Signed Ranks Test was administered to identify the significance of this difference. The test showed a significant difference in sleep quality between cannabis users and non-users, Z = -3.85, p = .000.

When isolating ‘Component 1’ of the PSQI, which is scored based on the question: “During the past month, how would you rate your sleep quality overall?”, no significant difference in subjective sleep quality was found between cannabis users who used cannabis as a sleep aid and non-users. Participants who used cannabis as a sleep aid had a mean score of 1.00 (range = 3.00), while non-users also had a median score of 1.00 (range = 3.00).

When isolating ‘Component 2’ of the PSQI, based on the question: “How long (in minutes) has it taken you to fall asleep each night?”, no significant difference in sleep onset latency was found between cannabis users who used cannabis as a sleep aid and non-users. Participants who used cannabis as a sleep aid had a mean score of 1.00 (range = 3.00), while non-users had a mean score of a 1.50 (range = 3.00). However, a Wilcoxon Signed Ranks Test showed no significant difference between these scores, Z = -1.670, p = .095.

4. Students who use cannabis as a sleep aid experience worse depression symptoms than non-users

Student cannabis users who used cannabis as a sleep aid, on average, experienced more symptoms of depression than non-users. Participants who used cannabis as a sleep aid had a median HADS depression score of 4.50 (range = 10.00), while non-users had a median HADS depression score of 3.00 (range = 19.00). A Wilcoxon Signed Ranks Test showed a significant difference in HADS depression scores between cannabis users and non-users, Z = -3.941, p = .000.

Discussion

The main findings of this study demonstrate no significant correlations between symptoms of depression and problematic cannabis use within this student sample, or between sleep quality and problematic cannabis use, despite the available literature providing alternative findings and conclusions. In terms of using cannabis as a sleep aid, a significant difference was found: non-users of cannabis had a better overall sleep quality than cannabis users. Also, a significant difference in depression symptom scores was found: non-users of cannabis had fewer or less severe symptoms of depression than cannabis users.

Contrary to the study completed by Babson, Boden and Bonn-Miller (2013), this study refuted the hypothesis that individuals with the most severe symptoms of depression (highest HADS depression scorers) had the most problematic cannabis use. The first hypothesis, “students with the most severe symptoms of depression have the most problematic cannabis use” did not produce any significant results, it was rejected. This means that there is no clear evidence here to suggest that problematic cannabis use influences symptoms of depression, which lends support to the idea that cannabis use and depression are not significantly related. Perhaps if the sample size had been larger, there may have been more participants with ‘moderate’ and ‘severe’ HADS scores, which would have allowed for HADS to be further tested in relation to the CUDIT-R and PSQI.

The second hypothesis, “students with the worst sleep quality have the most problematic cannabis use” was initially rejected, as there was no significant correlation between overall sleep quality (PSQI scores) and problematic cannabis use (CUDIT-R scores). This insignificant finding refutes the hypothesis that there is a relationship between sleep quality and cannabis use. Scientifically speaking, pre-sleep cannabis use does in fact decrease overall sleep quality, however the initial correlation between PSQI scores and CUDIT-R scores did not represent this. Had this correlation been significant, results would have supported the hypothesis that students with the worst sleep quality have the most problematic cannabis use. This would have supported relevant findings from the aforementioned study conducted by Freemon (1974) regarding pre-sleep cannabis use and its detrimental effects on overall sleep quality.

Moving on to the phenomenon of using cannabis as a sleep aid, the third hypothesis, “students who use cannabis as a sleep aid have the best self-reported sleep quality”, the initial result refutes the hypothesis, as non-users of cannabis actually showed a better sleep quality than participants who use cannabis as a sleep aid. The difference in PSQI scores between these two participant datasets was significant, meaning that in this student sample, non-users of cannabis had a better overall sleep quality than cannabis users. Also, when specifically looking at self-reported sleep quality by isolating ‘Component 1’ of the PSQI questionnaire, “During the past month, how would you rate your sleep quality overall?”, cannabis users who used cannabis as a sleep aid scored the same on self-reported sleep quality when compared to non-users, on average. This finding refutes previous research conducted by Babson, Boden & Bonn-Miller (2013), whose study supported the outcome that pre-sleep cannabis use improves perceived sleep quality. It is interesting to observe that this outcome is inconsistent when comparing this student sample to Babson, Boden and Bonn-Miller’s sample of medical cannabis users. When isolating ‘Component 2’ of the PSQI questionnaire, “How long (in minutes) has it taken you to fall asleep each night?”, no significant difference in sleep onset latency was found between cannabis users who used cannabis as a sleep aid and non-users. It may be necessary to examine these differences using a larger sample of student cannabis users and non-users among the general population, which would further paint the picture of how cannabis use impacts self-reported sleep quality and overall sleep quality. However in order for such a comparison to occur, a similar study would need to be conducted with a sample size larger than 38 participants who use cannabis as a sleep aid.

Finally, exploring the relationship between pre-sleep cannabis use and depression also produced significant results. The fourth hypothesis, “students who use cannabis as a sleep aid experience worse depression symptoms than non-users”, was supported by the significant difference found in HADS depression scores when non-users were compared with cannabis users who use cannabis as a sleep aid. Because cannabis users scored higher than non-users on the HADS depression questionnaire, this can be translated as non-users having a fewer or less severe symptoms of depression than cannabis users in this student sample. This finding is in line with the previous literature (Degenhardt, Hall & Lynskey, 2003) regarding this relationship. Further research should attempt to examine the reasons behind such a significant relationship, in other words; why using cannabis as a sleep aid or using cannabis in general is related to experiencing more symptoms of depression when compared to non-users. Previous research done, for example, by Bovasso (2001) suggests that individuals who use cannabis problematically are four times more likely to develop depressive symptoms in the future. Figuring out where using cannabis as a sleep aid fits into this relationship would provide useful insights into the complex relationship between depression, sleep quality and cannabis use.

It is curious to examine that even though the first hypothesis, “students with the most severe symptoms of depression have the most problematic cannabis use”, was rejected, the fourth hypothesis was accepted as significant. This means that even though students who use cannabis as a sleep aid do have significantly more symptoms of depression than non-users, students who use cannabis in a problematic manner do not experience more symptoms of depression than non-users. This finding could support the idea that cannabis users are using cannabis as a sleep aid because they cannot sleep without it due to their symptoms of depression. Student cannabis users may be using cannabis to help them sleep because depression is not allowing them to enter into a sleep state without the relaxing and arguably emotionally numbing nature of some strains of cannabis. However, the fact that hypotheses 1 and 4 are similar in direction but have produced completely opposite results suggests that further research should be done to establish exactly how pre-sleep cannabis use affects depression and whether this should be considered problematic or not.

Further analysis should be conducted on the cut-off point for problematic cannabis use in terms of the CUDIT-R. The study conducted by Annaheim, Rehm and Gmel (2008) suggested a cut-off point somewhere between 6 and 8 points; further research should be done to pinpoint an exact score. Furthermore, inclusion of a further revised CUDIT scale would be useful, as seen in a study conducted by Adamson et al. (2010), wherein a revised 8-item version of the CUDIT was constructed. This version was deemed to be an improvement in terms of internal consistency, discriminant validity, test-retest reliability, a one factor solution compared to the original two factor solution, as well as having a significant clinical utility, not just to identify cases (Adamson et al, 2010). Use of this version of the CUDIT-R may produce more accurate results in distinguishing different levels of problematic cannabis use, cannabis use disorders and the stages of change involved.

A further limitation was that of missing data values within the CUDIT-R scores. A number of participants did not fully complete the CUDIT-R; it cannot be known whether participants did not complete certain questions because they assumed that a lack of completion was equivalent to a score of ‘0’, or whether participants mistakenly failed to complete said questions. The intuitive idea to convert all missing CUDIT-R item scores to ‘0’ was henceforth rejected. In place of this, the SPSS function ‘Multiple Imputation’ for missing data was used to generate scores for missing items, based on all other completed items for that participant. While this generated seemingly representative scores, a true representation of participants’ responses could not have been known. This issue negatively affects confidence in the presented cannabis-related findings, however the extent of this effect is marginal because a fair replacement for missing data was implemented, both to increase the sample size of student cannabis users and to aid the creation of interesting findings. To dismiss a participant’s entire CUDIT-R score simply due to one incomplete item score would be wasteful; Multiple Imputation was therefore a viable option in terms of preserving valuable participant data.

Appendices

Abbildung in dieser Leseprobe nicht enthalten

Appendix 1: Participant Information Sheet

Abbildung in dieser Leseprobe nicht enthalten

Appendix 1: Participant Information Sheet

Abbildung in dieser Leseprobe nicht enthalten

Appendix 2: Demographics Questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Appendix 3: Substance Use Questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Appendix 3: Substance Use Questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Appendix 4: CUDIT-R questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Appendix 5: PSQI questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Appendix 5: PSQI questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Appendix 5: PSQI questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Appendix 6: HADS questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Appendix 6: HADS questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Abbildung in dieser Leseprobe nicht enthalten

Appendix 6: HADS questionnaire

Abbildung in dieser Leseprobe nicht enthalten

Appendix 7: List of relevant helplines and services

References

Green, B., Kavanagh, D., & Young, R. (2004). Reasons for cannabis use in men with and without psychosis. Drug And Alcohol Review, 23 (4), 445-453. http://dx.doi.org/10.1080/09595230412331324563

Bonn-Miller, M., Babson, K., & Vandrey, R. (2014). Using cannabis to help you sleep: Heightened frequency of medical cannabis use among those with PTSD. Drug And Alcohol Dependence, 136, 162-165. http://dx.doi.org/10.1016/j.drugalcdep.2013.12.008

Freemon, F. (1974). The effect of δ9-tetrahydrocannabinol on sleep. Psychopharmacologia, 35 (1), 39-44. http://dx.doi.org/10.1007/bf00421564

Cousens, K. & DiMascio, A. (1973). δ9 THC as an hypnotic. Psychopharmacologia, 33 (4), 355-364. http://dx.doi.org/10.1007/bf00437513

Chait, L. (1990). Subjective and behavioral effects of marijuana the morning after smoking. Psychopharmacology, 100 (3), 328-333. http://dx.doi.org/10.1007/bf02244601

Nicholson, A., Turner, C., Stone, B., & Robson, P. (2004). Effect of Delta-9-Tetrahydrocannabinol and Cannabidiol on Nocturnal Sleep and Early-Morning Behavior in Young Adults. Journal Of Clinical Psychopharmacology, 24 (3), 305-313. http://dx.doi.org/10.1097/01.jcp.0000125688.05091.8f

Babson, K., Boden, M., & Bonn-Miller, M. (2013). Sleep Quality Moderates the Relation between Depression Symptoms and Problematic Cannabis Use among Medical Cannabis Users. The American Journal Of Drug And Alcohol Abuse, 39 (3), 211-216. http://dx.doi.org/10.3109/00952990.2013.788183

Bovasso, G. (2001). Cannabis Abuse as a Risk Factor for Depressive Symptoms. American Journal Of Psychiatry, 158 (12), 2033-2037. http://dx.doi.org/10.1176/appi.ajp.158.12.2033

Deas, D. & Brown, E. (2006). Adolescent Substance Abuse and Psychiatric Comorbidities. The Journal Of Clinical Psychiatry, 67 (07), e02. http://dx.doi.org/10.4088/jcp.0706e02

Lopes, E., Milheiro, I., & Maia, A. (2013). Sleep quality in college students: a study about the contribution of lifestyle, academic performance and general well-being. Sleep Medicine, 14, e185. http://dx.doi.org/10.1016/j.sleep.2013.11.437

Aronin, S. & Smith, M. (2016). YouGov | One in four students suffer from mental health problems. YouGov: What the world thinks. Retrieved from https://yougov.co.uk/news/2016/08/09/quarter-britains-students-are-afflicted-mental-hea/

McGill University. (2007, October 24). Cannabis: Potent Anti-depressant In Low Doses, Worsens Depression At High Doses. ScienceDaily. Retrieved from www.sciencedaily.com/releases/2007/10/071023183937.htm

Aizpurua-Olaizola, O., Soydaner, U., Öztürk, E., Schibano, D., Simsir, Y., & Navarro, P. et al. (2016). Evolution of the Cannabinoid and Terpene Content during the Growth of Cannabis sativa Plants from Different Chemotypes. Journal Of Natural Products, 79 (2), 324-331. doi:10.1021/acs.jnatprod.5b00949

Neuroscience of sleep. (2017). En.wikipedia.org. Retrieved from https://en.wikipedia.org/wiki/Neuroscience_of_sleep#REM_sleep_activity

Degenhardt, L., Hall, W., & Lynskey, M. (2003). Exploring the association between cannabis use and depression. Addiction, 98 (11), 1493-1504. doi:10.1046/j.1360-0443.2003.00437.x

Adamson, S., Kay-Lambkin, F., Baker, A., Lewin, T., Thornton, L., Kelly, B., & Sellman, J. (2010). An improved brief measure of cannabis misuse: The Cannabis Use Disorders Identification Test-Revised (CUDIT-R)☆. Drug And Alcohol Dependence, 110 (1-2), 137-143. doi:10.1016/j.drugalcdep.2010.02.017.

Snaith, R. and Zigmond, A. (1986). The hospital anxiety and depression scale. BMJ, 292(6516), pp.344-344.

Buysse, D., Reynolds, C., Monk, T., Berman, S. and Kupfer, D. (1989). The Pittsburgh sleep quality index: A new instrument for psychiatric practice and research. Psychiatry Research, 28(2), pp.193-213.

Annaheim, B., Rehm, J. and Gmel, G. (2008). How to Screen for Problematic Cannabis Use in Population Surveys. European Addiction Research, 14(4), pp.190-197.

Adamson, S., Kay-Lambkin, F., Baker, A., Lewin, T., Thornton, L., Kelly, B. and Sellman, J. (2010). An improved brief measure of cannabis misuse: The Cannabis Use Disorders Identification Test-Revised (CUDIT-R)☆. Drug and Alcohol Dependence, 110(1-2), pp.137-143.

Health & Social Care Information Centre (2016). Statistics on Drug Misuse England, 2016. [online] Health & Social Care Information Centre, p.18. Available at: http://content.digital.nhs.uk/catalogue/PUB21159/drug-misu-eng-2016-rep.pdf

Lev-Ran, S., Roerecke, M., Le Foll, B., George, T., McKenzie, K. and Rehm, J. (2014). 892 – The association between cannabis use and depression: a systematic review and meta-analysis of longitudinal studies. European Psychiatry, 28, p.1.

Harder, V., Morral, A. and Arkes, J. (2006). Marijuana use and depression among adults: testing for causal associations. Addiction, 101(10), pp.1463-1472.

Smyth, C. (2012). The Pittsburgh Sleep Quality Index (PSQI). ‘try this:’ Best Practices in Nursing Care to Older Adults, 6(1). Available at: https://consultgeri.org/try-this/general-assessment/issue-6.1.pdf

Bocéréan, C. and Dupret, E. (2014). A validation study of the Hospital Anxiety and Depression Scale (HADS) in a large sample of French employees. BMC Psychiatry, 14(1).

[...]