Please provide R code for each Question.

Chronic conditions are diseases that last one year or more and require ongoing medical attention,

such as diabetes and cancer. The growing number of patients with chronic conditions has significant

economic implications; according to the Centers for Disease Control, diabetes costs the

United States health care system and employers $237 billion annually.

A strategy for reducing the burden on health care systems is promoting patient self-management,

in which patients proactively take action to maintain and improve their own health. For patients

with a chronic condition, daily maintenance constitutes a form of prevention that can lead to better

quality of life and reduce health care costs. In individuals with diabetes, for instances, adhering

to recommended diet and exercise regimens can help reduce the risk of serious complications

like heart disease or stroke.

Self-management is not appropriate for all patients. A study was conducted among 1,154 adult

patients from different regions in the Netherlands to identify characteristics associated with activation

for self-management. Study participants had a clinical diagnosis of one of four chronic

conditions: Chronic Obstructive Pulmonary Disease (COPD), Chronic Heart Failure (CHF), Diabetes

Mellitus Type II (DM-II), or Chronic Renal Disease (CRD).

The primary outcome, activation for self-management, was assessed with the Patient Activation

Measure (PAM-13), a questionnaire which assesses self-reported knowledge, skills, and confidence

for self-management. Score on the PAM-13 ranges from 0 to 100, with a higher score indicating

a higher level of self-management activation. Scores can be classified into one of four

levels: Level 1 ( 47:0 points), Level 2 (47:1 ???? 55:1 points), Level 3 (55:2 ???? 67 points), and Level 4

( 67:1 points). A Level 1 patient demonstrates lack of knowledge and confidence to manage their

condition. A Level 2 patient demonstrates a limited amount of knowledge and ability to set goals.

Both Level 1 and Level 2 patients typically believe that their health is largely out of their control

and are considered passive recipients of care. In contrast, a Level 3 patient is goal-oriented and

actively working to achieve best practice behaviors. A Level 4 patient has successfully maintained

new behavior over time.

In addition to the PAM-13, participants completed assessments measuring health status, anxiety

and depression, and perceived social support.

- Health Status. Higher scores on the Short Form-12 Health Survey (SF-12) are indicative

of better health status; possible scores range from 0-100 points. The survey consists of

two summary scores on the 0-100 scale, one measuring physical health and one measuring

mental health.

- Anxiety and Depression. Higher scores on the Hospital Anxiety and Depression Scale (HADS)

indicates a higher state of anxiety or depression. Scores range from 0-21 points for each

component, and a score of 11 or greater suggests the presence of an anxiety or depressive

disorder.

- Social Support. Higher scores on the Multidimensional Scale of Perceived Support (MSPSS)

indicates higher perceived support, such as from family, friends, and significant others.

Scores range from 0-84 points.

Information was also collected on participant sociodemographics, such as level of financial distress

and educational level. Educational level was divided into lower (primary school through

vocational training), middle (secondary school), and high (college or university) education, according

to the Dutch school system.

Data from the study are in the file self_manage.Rdata. The following table provides a list of the

variables in the dataset and their descriptions.

Variable Description

sex sex, coded female or male

height height in centimeters (cm)

weight weight in kilograms (kg)

financial financial distress, either none, low, or high

age age in years

bmi body mass index (kg/m2)

pam.score PAM-13 score, on 0-100 point scale

pam.cat PAM level, from Level 1 to Level 4

sf.total SF-12 total score, on 0-100 point scale

sf.phys SF-12 score for physical health, on 0-100 point scale

sf.ment SF-12 score for mental health, on 0-100 point scale

has.depress HADS score for depression, on 0-21 point scale

has.anxiety HADS score for anxiety, on 0-21 point scale

supp.total MSPSS score for perceived social support, on 0-84 point scale

disease chronic disease, either DM-II, COPD, HF, or CRD

edu educational level, either lower, middle, or higher

duration disease duration

severity disease severity, either mild, moderate, or severe

Use the data to answer the following questions.

a) (6 pts.) Provide an informative summary describing features of the study participants with respect

to the variables age, sex, and type of chronic disease. Reference appropriate numerical

and graphical summaries as needed. Limit your answer to no more than five sentences.

b) (6 pts.) The main measurement of interest is activation for self-management. Describe the

distribution of activation for self-management within the study participants, both in terms

of PAM-13 score and PAM level. Reference appropriate numerical and graphical summaries

as needed. Limit your answer to no more than five sentences.

c) (12 pts.) Explore the relationship between activation for self-management and disease type.

i. provide a plot to graphically show the association between disease Provide and PAM-13

score. Describe what you see.

ii. Provide a summary that shows how the distribution of PAM level differs between disease

type. Describe what you see.

iii. Do you find the summary from part i. or the summary from part ii. more informative

with regards to understanding the relationship between activation for self-management

and disease type? Explain your answer in no more than five sentences.

d) (10 pts.) Is PAM-13 score associated with perceived level of social support?

i. Using graphical and numerical methods, investigate this question separately within

disease types. Summarize your findings, with particular focus on whether the association

between PAM-13 score and perceived level of social support seems to differ

between types of chronic disease.

ii. Do these data suggest that an increase in (perceived) social support leads to better

capacity for self-management? Explain your answer.

e) (8 pts.) Investigate the relationship between PAM-13 score and educational level.

i. With reference to appropriate numerical and graphical summaries, describe the association

between PAM-13 score and educational level.

ii. Propose one possible explanation for the trends observed in part i. Limit your answer

to no more than five sentences.

f) (14 pts.) Investigate the relationship between PAM level, age, and educational level.

i. Provide a graphical summary that shows the association between age and PAM level.

Describe what you see.

ii. Create graphical summaries that show the association between age and PAMlevel when

comparing individuals of the same educational level. Describe what you see.

iii. A news outlet is interested in reporting on the study results. You have been asked to

address the following question: "Do these data suggest that older individuals tend to

have a lower level of activation for self-management?"

Based on the findings from parts i. and ii., address the question in language accessible

to a non-statistician. Limit your answer to no more than five sentences.

g) (16 pts.) Explore the relationship between depression and activation for self-management.

i. From these data, compare the risk (i.e., probability) of being PAM Level 1 for individuals

classified as having depressive disorder versus those not classified as having a

depressive disorder.

ii. Calculate the difference in mean PAM-13 score between individuals classified as having

a depressive disorder versus those not classified as having a depressive disorder. In one

sentence, report the calculation; use phrasing that is informative for a general audience.

iii. There are some individuals missing responses for the HADS questionnaire. Does this

missingness represent a potential source of bias for the calculations in parts i. and ii?

Explain your answer.