For those with insomnia, the Athens Insomnia Scale (AIS) will be used to assess insomnia symptoms. This scale ranges from 0-24, with higher scores indicating worse sleep outcomes. Reported results are an average taken over weeks 2, 4, and 12 to be consistent with the regression model estimates.

The Short Form 12 Health Survey (SF-12) Mental Health Summary Scale was used to assess self-report of mental health. SF-12 scores are based on an Item Response Theory model, in which they are z-scores rescaled to have a mean of 50 (SD=10) in any typical population. This measure has no range limits. Higher scores indicate better mental health functioning. Reported results are an average taken over weeks 2, 4, and 12 to be consistent with the regression model estimates.

Congruency cost was assessed wth the Attention Switching Task (AST), which assesses attention shifting and executive function via a simple task assessing the direction (pointing left or right) and position (on the left or right side) of an arrow, and is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range of this measure is -1900 to 1900 ms. It is calculated as the difference in the median latency for incongruent trials (i.e., trials where the direction of the arrow differs from the side it appears) minus the median latency for congruent trials (i.e., trials where the direction of the arrow matches the side it appears). Positive values indicate faster responses to congruent trials. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

Switching cost was assessed with the Attention Switching Task (AST), which assesses attention shifting and executive function via a simple task assessing the direction (pointing left or right) and position (on the left or right side) of an arrow, and is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range for this measure is -1900 to 1900 ms. It is calculated as the difference in the median latency for blocks of switching trials (i.e., blocks of trials where the response rule switches between indicating the direction of the arrow or the position of the arrow) minus the median latency for blocks of non-switching trials (i.e., blocks of trials where the response rule is always either to indicate the direction of the arrow or the position of the arrow across all trials). Positive scores indicate faster responses to non-switching trials. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

Discriminability was assessed with the Rapid Visual Information Processing (RVP) Task, which assesses sustained attention and is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). Discriminability scores are based on a transformation of response times and accuracy using the EZ-Diffusion model. There are no limits for this measure. Higher scores indicate that subjects can better discriminator between targets and distractors. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

Total errors were assessed with the Paired Associates Learning (PAL) task, which assesses visual memory and is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range for this measure is 0 to 70 errors. Higher scores indicate that the subject committed more errors (worse performance). Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

Repetition errors were assessed with the Spatial Working Memory (SWM) task, which assesses spatial working memory and executive function. It is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range for this measure is 0 to 63 errors. Higher scores indicate that more repetition errors were committed (worse performance). Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

Discriminability was assessed with the Verbal Recognition Memory task (VRM), which assess verbal memory. It is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). Discriminability scores are based on a transformation of hit and false alarm rates using a signal detection theory (SDT) model.There are no limits for this measure. Higher scores indicate that a subject was able to better discriminate between previously studied and novel words. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

Free Recall Memory was assessed with the Verbal Recognition Memory task (VRM), which assesses verbal memory. It is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range for this measure is 0 to 18 words recalled. Higher scores indicate that subjects were able to correctly recall more previously studied words. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

Total of all reporting groups

This group agrees to wait 3 months before using medical marijuana.

For those with depression symptoms, the depression subscale of the Hospital Anxiety and Depression Scale (HADS) will be used to assess symptoms. The subscale ranges from 0-21, with a higher score indicating worse depression outcomes. Reported results are an average taken over weeks 2, 4, and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 83 participants out of 186 total who intended to use medical marijuana to address concerns for anxiety and depression.

For those with anxiety symptoms, the anxiety subscale from the Hospital Anxiety and Depression Scale (HADS) will be used to assess symptoms. This subscale ranges from 0-21, with higher scores indicating worse anxiety outcomes. Reported results are an average taken over weeks 2, 4, and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 83 participants out of 186 total who intended to use medical marijuana to address concerns for anxiety and depression.

For those with pain, the Brief Pain Inventory (BPI) severity subscale was used to assess pain symptoms. This scale ranges from 0-10, with a higher score indicating greater pain severity. Reported results are an average taken over weeks 2, 4, and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 61 participants out of 186 total who intended to use medical marijuana to address concerns for pain.

For those with insomnia, the Athens Insomnia Scale (AIS) will be used to assess insomnia symptoms. This scale ranges from 0-24, with higher scores indicating worse sleep outcomes. Reported results are an average taken over weeks 2, 4, and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 42 participants out of 186 total who intended to use medical marijuana to address concerns for insomnia.

The Short Form-12 Health Survey (SF-12) Physical Health Summary Scale was used to assess self-report of physical health. SF-12 scores are based on an Item Response Theory model, in which they are z-scores rescaled to have a mean of 50 (SD=10) in any typical population. There are no range limits to this measure. Higher scores indicate better physical health functioning. Reported results are an average taken over weeks 2, 4, and 12 to be consistent with the regression model estimates.

The Short Form 12 Health Survey (SF-12) Mental Health Summary Scale was used to assess self-report of mental health. SF-12 scores are based on an Item Response Theory model, in which they are z-scores rescaled to have a mean of 50 (SD=10) in any typical population. This measure has no range limits. Higher scores indicate better mental health functioning. Reported results are an average taken over weeks 2, 4, and 12 to be consistent with the regression model estimates.

Congruency cost was assessed wth the Attention Switching Task (AST), which assesses attention shifting and executive function via a simple task assessing the direction (pointing left or right) and position (on the left or right side) of an arrow, and is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range of this measure is -1900 to 1900 ms. It is calculated as the difference in the median latency for incongruent trials (i.e., trials where the direction of the arrow differs from the side it appears) minus the median latency for congruent trials (i.e., trials where the direction of the arrow matches the side it appears). Positive values indicate faster responses to congruent trials. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 162 participants out of 186 total who completed the AST module of the CANTAB at weeks 4 and 12.

Switching cost was assessed with the Attention Switching Task (AST), which assesses attention shifting and executive function via a simple task assessing the direction (pointing left or right) and position (on the left or right side) of an arrow, and is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range for this measure is -1900 to 1900 ms. It is calculated as the difference in the median latency for blocks of switching trials (i.e., blocks of trials where the response rule switches between indicating the direction of the arrow or the position of the arrow) minus the median latency for blocks of non-switching trials (i.e., blocks of trials where the response rule is always either to indicate the direction of the arrow or the position of the arrow across all trials). Positive scores indicate faster responses to non-switching trials. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 162 participants out of 186 total who completed the AST module of the CANTAB at weeks 4 and 12.

Discriminability was assessed with the Rapid Visual Information Processing (RVP) Task, which assesses sustained attention and is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). Discriminability scores are based on a transformation of response times and accuracy using the EZ-Diffusion model. There are no limits for this measure. Higher scores indicate that subjects can better discriminator between targets and distractors. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 177 participants out of 186 total who completed the RVP module of the CANTAB at weeks 4 and 12.

Total errors were assessed with the Paired Associates Learning (PAL) task, which assesses visual memory and is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range for this measure is 0 to 70 errors. Higher scores indicate that the subject committed more errors (worse performance). Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 177 participants out of 186 total who completed the PAL module of the CANTAB at weeks 4 and 12.

Repetition errors were assessed with the Spatial Working Memory (SWM) task, which assesses spatial working memory and executive function. It is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range for this measure is 0 to 63 errors. Higher scores indicate that more repetition errors were committed (worse performance). Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 177 participants out of 186 total who completed the ? module of the CANTAB at weeks 4 and 12.

Discriminability was assessed with the Verbal Recognition Memory task (VRM), which assess verbal memory. It is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). Discriminability scores are based on a transformation of hit and false alarm rates using a signal detection theory (SDT) model.There are no limits for this measure. Higher scores indicate that a subject was able to better discriminate between previously studied and novel words. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 171 participants out of 186 total who completed the VRM module of the CANTAB at weeks 4 and 12.

Free Recall Memory was assessed with the Verbal Recognition Memory task (VRM), which assesses verbal memory. It is part of the Cambridge Neuropsychological Test Automated Battery (CANTAB). The range for this measure is 0 to 18 words recalled. Higher scores indicate that subjects were able to correctly recall more previously studied words. Reported results are an average taken over weeks 4 and 12 to be consistent with the regression model estimates.

The analysis is restricted to the subset of 175 participants out of 186 total who completed the free recall VRM module of the CANTAB at weeks 4 and 12.

This group agrees to wait 3 months before using medical marijuana.