Roy's blog

Data Management and Visualization by Wesleyan University - Assignment(Week 1): Developing a Research Question and Creating a Personal Code Book

Introduction:

After reviewing the codebook of NESARC, a survey of over  43093 U.S. citizens (over the age of 18) designed to determine the magnitude of alcohol use and psychiatric disorders, I found myself particularly interested in cannabis use disorders. More specifically, I aim to examine the patterns of association between cannabis use, major depression, and general anxiety disorder diagnosed in the last 12 months. Nowadays, cannabis is the most widely consumed drug in many countries. However, there is also a medical use of it. It has been estimated that around 10% of people who use cannabis will depend on it. [4] It is common knowledge that there is a strong relationship between mental disorders and alcohol abuse. Therefore, it is interesting whether cannabis’ frequent use co-occurs with depression and generalized anxiety.

Research Question:

Is cannabis use associated with major depression and general anxiety disorder diagnoses in the last 12 months?

Hypothesis:

Despite the fact that several studies have examined issues relevant to cannabis abuse disorders, it's difficult to decide if cannabis use leads to psychiatric illnesses. At this point, it is necessary to separate cannabis use from a heavier involvement with the drug (dependence/abuse). My personal belief is that cannabis use increases the likelihood of depression symptoms and anxiety disorders, however not as significantly as cannabis abuse/dependence, which could cause mental disorders like the ones mentioned above.

NESARC Codebook Sections and Variables:

After looking through the NESARC codebook, firstly I decided to take into consideration the unique identification number ( IDNUM ) and the variable (AGE) from the background information ( SECTION 1 ) of the sample, in order to make my findings more reliable. Furthermore for the first question topic I chose, from the drug/medicine use section ( SECTION 3B ), to include information like the percentage of people who ever used cannabis ( S3BQ1A5 ), as well as the period of this use -last 12 months / prior to last 12 months / both periods- ( S3BD5Q2B ) and the frequency of it when using the most ( S3BD5Q2E ). As far as the second topic is concerned, I selected the variable of non-hierarchical major depression diagnoses, in the last 12 months ( MAJORDEP12 ) and the variable of non-hierarchical generalized anxiety diagnoses, in the last 12 months ( GENAXDX12 ), which are included in the diagnoses section ( SECTION 14 ).

SECTION 1  variables : IDNUM ,  AGE

SECTION 3B variables:S3BQ1A5 ,S3BD5Q2B ,  S3BD5Q2E

SECTION 14 variables:  MAJORDEP12,GENAXDX12

Literature review:

Considering the literature review I performed using  Google Scholar, I found several academic studies and research on the relationship between cannabis use, depression, and anxiety. There was a moderate association between involvement with cannabis use in the past 12 months and the prevalence of affective and anxiety disorders. Among those with DSM-IV cannabis dependence, 14 % had affective disorder symptoms, compared to 6 % of non-users, while 17 % met the criteria for an anxiety disorder, compared to 5 % of non-users [4]. After including demographics, neuroticism, and other drug use in multiple regressions, these associations did not remain significant. Cannabis use did not appear to be directly related to depression or anxiety when the account was taken of other drug use. However, the association between heavier involvement with cannabis use and affective and anxiety disorders has implications for treating persons with problematic cannabis use [4].      

References:

Degenhardt, L., Hall, W., & Lynskey, M. (2003). Exploring the association  between cannabis use and depression. Addiction, 98(11), 1493-1504.

Hayatbakhsh, M. R., Najman, J. M., Jamrozik, K., Mamun, A. A., Alati, R., & Bor, W. (2007). Cannabis and anxiety and depression in young adults: a large prospective study. Journal of the American Academy of Child & Adolescent Psychiatry, 46(3), 408-417.

Degenhardt, L., Hall, W., & Lynskey, M. (2001). Alcohol, cannabis and tobacco use among Australians: a comparison of their associations with other drug use and use disorders, affective and anxiety disorders, and psychosis. Addiction, 96(11), 1603-1614.

Let's say that you are interested in studying the relationship between exercise and heart health. You want to investigate how different variables might affect this relationship.

Mediator: One potential mediator in this example is cholesterol levels. High levels of cholesterol are associated with an increased risk of heart disease, and it's possible that exercise could help to lower cholesterol levels, which could in turn lead to better heart health. Cholesterol levels would be a mediating variable that explains the relationship between exercise and heart health.

Moderator: One potential moderator in this example is age. It's possible that the relationship between exercise and heart health might be stronger for older adults, who are at a higher risk for heart disease, compared to younger people. Age would be a moderating variable that affects the strength or direction of the relationship between exercise and heart health.

Confounding: A potential confounding variable in this example could be diet. People who exercise regularly may also be more likely to eat a healthy diet, and a healthy diet is also associated with better heart health. This could lead to the erroneous conclusion that exercise is associated with better heart health when it is actually diet that is driving the relationship.

Control: Finally, a potential control variable in this example could be smoking status. Smoking is a well-known risk factor for heart disease, and it's possible that people who exercise regularly are also less likely to smoke. By controlling for smoking status, you can ensure that any differences in heart health are not due to differences in smoking behavior.

As a Data Analyst, it's important to consider all of these different variables when conducting a study. By understanding the role of mediator, moderator, confounding, and control variables, we can make more accurate conclusions about the relationships between different variables and better understand the factors that influence health outcomes.