mrssaw

[mrssaw]

Creating graphs

The assignment was to create univariate graphs on the one hand, and bivariate graphs on the other hand.

Univariate graphs for each selected variables:

MAJORDEP12 and AGE

2. S4AQ4A14

3. MAJORDEP12 and S4AQ4A14

4. S4AQ54

5. MAJORDEP12 and S4AQ54

6. MAR12ABDEP

7. S4AQ6B

8. MAR12ABDEP and S4AQ6B

Bivariate graphs:

Summary:

As the variables were all categorical, there was no possibility to calculate the center and spreads

the graphs showed that the 20 years old people have the highest number of major depression compared to the other young adults between 18 and 30 years old. Although there are some peaks, the number of major depression among young adults seems to be equally distributed and there is no tendency to be seen

thy symptom of having trouble to concentrate/keeping mind on things among young people with major depression shows no direct association, as there are less people with that symptom compared to the total number of people (all ages)

however, the symptom of having trouble doing things supposed to do is a bit more widespread among young people. But compared to the entirety it is not strongly pronounced

the graphs revealed that there is a very low number of people with any kind of cannabis diagnosis and also a low number of young adults whose had an onset of first episode in the last 12 months

the fewest people who had their onset of first episode had a cannabis diagnosis.

some had an onset during cannabis abuse/dependence. but, the majority of those who consumed cannabis in the last 12 months had no onset

[mrssaw]

Data Management decisions

As next step of the course I have managed the data of my first program.

Missing data that was declared as "9: Unknown" I have replaced to NaN

Some Codes had different variables with the same meaning, e.g. on the one hand one code said 1=yes and 2=no, on the other hand one code said 1=yes and 0=no. That could be confusing. So I recoded the variables, so that in the results only yes and no are displayed.

S4AQ4A14

2. MAJORDEP12

3. S4AQ6B

As you can see in the results, there is sometimes a huge amount of missing data. Therefore, it is important to declare them as NaN to prevent any misunderstanding.

The replacement of the numbers to "yes"/"no" makes it more easier to understand the figures.

[mrssaw]

My first Program

I wrote a program examining two research questions.

How is the occurrence of symptoms associated with the onset of episodes?

What impact do drugs have on the onset of episodes?

Here, I focused on young people at the age betwenn 18 and 30 years old

To answer the first question I used the codes:

MAJORDEP12 - Major depression in last 12 months (non-hierarchical)

AGE - the age of the people

S4AQ4A14 - Had trouble concentrating/keeping mind on things most days for 2+ weeks

S4AQ54 - Had trouble doing things supposed to do - like working, doing schoolwork, or taking care of home/family

To answer the second question I used these codes:

MAR12ABDEP - Cannabis abuse/dependence in last 12 months

S4AQ6B - Was onset of first episode in last 12 months

The program:

Program for Variable 1: MAJORDEP12

The program considers the number of people with major depression between 18 and 30 years old.

The output for Variable 1: MAJORDEP12

Program for Variable 2: S4AQ4A14

The program counts first for the number of young adults between 18 and 30 years old in regard of having trouble concentrating/keeping mind on things most days for 2+ weeks.

Then it counts for the number of people between 18 and 30 years old with that symptom and the occurrence of major depression.

Output of variable 2: S4AQ4A14

Program for variable 3: S4AQ54

The program counts first for the number of young adults between 18 and 30 years old in regard of having trouble doing things supposed to do - like working, doing schoolwork or taking care of home/family.

Then it counts for the number of people between 18 and 30 years old with that symptom and the occurrence of major depression.

Output for variable 3: S4AQ54

Program for variable 4: MAR12ABDEP

The program counts for the number of young adults between 18 and 30 years old in regard of cannabis abuse/dependence.

Additionally, it examines if there was an onset of first episodes in the last 12 months.

Output of variable 4: MAR12ABDEP

Program for variable 5: S4AQ6B

The program examines if there is an impact of cannabis on the onset of episodes.

First it considers only those who had their onset of the first episode in last 12 months and if there was any cannabis abuse/dependence.

Then, in the second step, it examines if those who had any kind of cannabis abuse/dependence an onset of a first episode within the last 12 months.

Output for variable 5: S4AQ6B

Conclusion

In the last 12 months 29.6% of the people with major depression were between 18 and 30 years old. The count was 1030. The mean average age was 23.7 years old. The standard deviation was 3.8 years.

About 21.2% of the young people between 18 and 30 years old had trouble concentrating/keeping mind on things for most days 2+ weeks. 9.37 % do not. It is needed to say that about 69.5% of the data ist missing. From the 2018 people with that symptom do 888 people have major depression (~44%).

13.18% of the young adults have to symptom of having trouble doing things they are supposed to do. 7.2% do not have this symptom. 79.6% did not answer to the question. 53.8% of the young people with major depression had trouble doing things they are supposed to do.

96% of the people between 18 and 30 years old had no cannabis diagnosis in the last 12 months. About 2.9% met the criteria 1: cannabis abuse only. Only 3.1% of the young people had their onset of first episode in the last 12 months- 17.77% did not have and 79% of the data is missing.

95.8% of the young people who had their onset of first episode in the last 12 months had no cannabis diagnosis. 6.6% of the people between 18 and 30 years old who consumed cannabis in the last 12 months had their first episode in the last 12 months as well. 34.64 did not have their onset of first episode and 58.75% of the data is missing.

To sum up, the occurrence of the symptom S4AQ54 is associated with the onset of major depression. The symptom S4AQ4A14 occurs also with major depression but not that often as the other symptom.

Regarding the second research question, there is no association between the consumption of cannabis and the onset of episodes within 12 months, as the number of people who consumed cannabis and got their first episodes is quite low compared to the others. However, it might be that it just takes more than 12 months until a depression caused by cannabis consumption is developed.

[mrssaw]

Research Project

Step 1: Choosing a data set

I have looked through the codebook for NESARC study and have decided that I want to know more about low mood/major depression.

How is the occurrence of symptoms associated with the onset of episodes?

Step 2: Identify a specific topic of interest

As I started studying for my master's degree and still continued working I had trouble doing things supposed to do and concentrating/keeping mind on things. I felt like that I getting into depression. I want to learn more about the reasons for the occurrence of major depression.

Therefore, I am interested in exploring the association between the symptoms like having trouble doing things to do and the onset of first episode in the last 12 months.

Step 3: Prepare codebook

As it is a very complex field, I am not sure yet which variables I should consider. Thus I will add to my codebook variables reflecting the symptoms and diagnosis.

Step 4: Identify a second topic

As I know some friends taking drugs that seem to have an effect on their mood, I want also to consider the association between drug use and diagnosis.

Step 5: Add questions/items/variables documenting this second topic to your personal codebook

I add to my codebook variables reflecting drug use.

What impact do drugs have on the onset of episodes?

Step 6: Literature review

In the article 'Drug Use and the Risk of Major Depressive Disorder, Alcohol Dependence, and Substance Use Disorders' (2022) by David W. Brook, et al the association between early drug use and later psychiatric disorders was investigated. The result was that early drug use was significantly related to later psychiatric disorders.

The paper 'Depression and Drug Use' (2010) by C. N. Stefanis, A. Kokkevi discuss the findings that there is a progressive increase of depressive mood levels is found as we proceed from the non-users to the users of different drugs/substances.

The study 'Comorbidity between DSM-IV drug use disorders and major depression: Results of a national survey of adults' (1995) by Bridget F. Gant describes detailed patterns of comorbidity between Diagnostic and Statistical Manual of Mental Disorders  drug use disorders and major depression in the United States for 1992. The result was that the association between drug dependence and major depression was greater than the association between abuse and major depression.

Step 7: Hypothesis

Drug use affects the probability of the occurrence of major depression.