Reference no: EM132623656
HMG7210 Biostatistics - Victoria University
Assessment
Learning outcome 1: Elaborate fundamental concepts in biostatistics in public health emphasizing the disciplines of nutrition and active living;
Learning outcome 2: Appraise the commonly used methods of analysis of data and exhibit specialised skills in handling data sets;
Learning outcome 3: Apply appropriate statistical methods to the design of a public health research study;
Learning outcome 4: Perform basic statistical analyses and present findings in a clear, concise and logical manner accessible to specialist and non-specialist audiences; and
Learning outcome 5: Critically evaluate statistical methods described and applied in global nutrition and active living literature in order to substantiate their potential contribution to the broader public health evidence-base.
Part 1 - Dyslipidaemia Study
For Part 1 you will need the dataset Dyslipidaemia.sav
A cross-sectional study was conducted on 366 diabetic patients to assess their blood lipid profile and to investigate possible factors contributing to the development of dyslipidaemia among diabetic patients. The patients were recruited from a specialized hospital outpatient department. All study participants provided a blood sample and completed a lifestyle and demographic questionnaire.
The objective of the study was to find factor(s) that may relate to the risk of developing dyslipidaemia in diabetic patients. Along with the demographic information and physical measurements, haematological measurements were reported on a continuous scale. For analysis purposes most of the continuous variables were transformed into grouped data. However, original continuous variables were also kept in the data set. The short and long names of the variables and corresponding statistical codes are presented in the following table, Table 1.
Question 1: Complete the column "Variable type", listing each of the variables as either nominal, ordinal, discrete or continuous.
Table 1: Variable description of the Dyslipidaemia study
|
Variable
|
Variable label
|
Value label
|
Variable type
|
ID
|
ID Number
|
NA
|
|
Age
|
age of patients
|
NA
|
|
Age_gr
|
age categories
|
1 < 40 years
2 40 - 49 years
3 50 - 59 years
4 >= 60 years
|
|
Gender
|
patients gender
|
1 Male
2 Female
|
|
Occupation
|
Occupation type
|
1 Sedentary worker
2 Moderate worker
3 Heavy physical worker
|
|
Physical_activity
|
Physical activity level
|
0 Sufficient PA
1 Insufficient PA
|
|
weight
|
weight of patients
|
NA
|
|
Height
|
height of patients
|
NA
|
|
BMI
|
body mass index of patients
|
NA
|
|
BMI_gr
|
body mass index category
|
0 Normal
1 Overweight
2 Obese
|
|
BP_diastolic
|
diastolic blood pressure
|
NA
|
|
BP_systolic
|
systolic blood pressure
|
NA
|
|
HTN
|
Hypertension
|
0 Normotensive
1 Hypertensive
|
|
Duration_diabetes
|
Duration of Diabetes
|
NA
|
|
Duration_diabetes_gr
|
Duration category
|
1 < 5 years
2 5 - <10 years
3 > = 10 years
|
|
HbA1c
|
Fasting HbA1c level
|
NA
|
|
HbA1c_gr
|
HbA1c categories
|
0 <7 Controlled
1 >=7 Uncontrolled
|
|
Insulin
|
Takes insulin
|
0 No
1 Yes
|
|
Duration_insulin
|
Duration of insulin (years)
|
NA
|
|
T.Chol
|
Total Cholesterol
|
NA
|
|
T_Chol_gr
|
Cholesterol category
|
0 < 200
1 >= 200
|
|
TG
|
Triglyceride
|
NA
|
|
TG_gr
|
Triglyceride category
|
0 < 150
1 >= 150
|
|
HDL
|
HDL Cholesterol
|
NA
|
|
HDL_gr
|
HDL category
|
0 Normal
1 Low
|
|
LDL
|
LDL Cholesterol
|
NA
|
|
LDL_gr
|
LDL category
|
0 < 100
1 >= 100
|
|
Creatinine
|
Serum Creatinine
|
NA
|
|
Createnine_gr
|
Creatinine category
|
0 < 1.5
1 >=1.5
|
|
TC_HDL
|
TC_HDL ratio
|
NA
|
|
LDL_HDL
|
LDL_HDL ratio
|
NA
|
|
Dyslipidemia
|
Dyslipidaemia
|
0 No dislipidemia
1 Dislipidemia
|
|
Question 2 Part a)Fill out the following table using appropriate summary statistics for each of the listed variables. The outcome variable is dyslipidaemia.
Variable
|
Dyslipidaemia (% or median & IQR or mean & SD)
|
Yes
|
No
|
HbA1c
|
|
|
BP diastolic
|
|
|
Insulin status:
|
|
|
Yes
|
|
|
No
|
|
|
Age group:
|
|
|
< 40 yrs.
|
|
|
40 - 49 yrs.
|
|
|
50 - 59
|
|
|
≥ 60 yrs.
|
|
|
Question 2 Part b) Discuss whether dyslipidaemia is related to HbA1c, diastolic BP, insulin status and patients' age.
- Evaluate the conditions (if any) that you should check before reporting summary statistics for each of the numerical variables in this question.
- Present the relevant analysis output below, along with the table/graph number and appropriate title for the table/graph (copy and paste all analysis output in a picture format by using "Copy Special").
- Discuss the results and make a brief summaryconclusion in the context of the study. Please present the discussion and conclusion separately.
Question 3: Graphically determine the relationship between dyslipidaemia and each of the following variables: duration of diabetes group, BMI group and occupation type. Present each of the above variables on an appropriate graph and discuss. In this task you will need to: a) Justify the choice of the graphs that have been used; b) Discuss the results and make a brief summary conclusion in the context of the study, and c) Present the graphs in the text along with the graph number and appropriate title for the graph (copy and paste all graphs in a picture format, see the instructions provided in Question 2 above)
Part 2-Cardiovascular Disease Study
For Part 2 you will need the dataset CVD-Data.sav
Consider a study where twelve baseline variables (e.g., age, gender, weight, BMI, etc.) among others were collected from each of the 1429 participants. Out of the 1429 participants, 952 had a history of being treated with Drug-X on a regular basis. The objective of the study was to investigate whether Drug-X was protective for various risk factors of cardiovascular vascular disease (CVD). The participants were then followed for 5 years and their mortality status was recorded. The short and long names of the variables are given in Table 2.
Table 2. Variable description for cardiovascular disease study
|
Variable
|
Description
|
Statistical code (if any)
|
AGE
|
Age
|
N/A
|
GENDER
|
Gender
|
0 for female and 1 for male
|
WEIGHT
|
Weight
|
N/A
|
BMI
|
Body mass index
|
0 for <25; 1 for 25 to <30; 2 for 30 to <35 and 3 for 35 and above
|
SBP
|
Systolic blood pressure
|
N/A
|
DBP
|
Diastolic blood pressure
|
N/A
|
CHL
|
Cholesterol level
|
N/A
|
SMOKE
|
Smoking status
|
0 for never smoked and 1 for smoked
|
Drug-X
|
Drug "X" treated
|
0 for no and 1 for yes
|
PAD
|
Peripheral arterial disease
|
0 for no and 1 for yes
|
STROKE
|
Stroke
|
0 for no and 1 for yes
|
DEATH
|
Follow-up mortality status
|
0 for no and 1 for yes
|
Question 4: Consider that an investigator aims to assess whether followed-up mortality (DEATH) was related to body mass index (BMI, a categorical variable) in the study sample. Evaluate the investigator's aim by presenting the related data on an appropriate graph. (note: no additional summary statistics or significance test required).
Question 5: Using an appropriate graphical presentation, discuss whether diastolic blood pressure (DBP) is related to body mass index (BMI, a categorical variable) in the study sample (note: no additional summary statistics or significance test is required). Based on the findings can you make an inference on the study population? Justify your answer.
Instructions for Questions 6 & 7:
- Justify the choice of the statistical methods that have been used,
- Present the data analysis output graphs/tables along with the graph/table number and appropriate title for the graph/table (please use the special copy and paste option when copying and pasting from SPSS),
- For statistical significance tests, please follow the steps that have been outlined in the lecture and tutorial slides. All steps should be in the context of the study. Note: Interpretation for some of the results are required.
- Please do not discuss the steps of significance tests for each of the variables separately. Instead combine SBP, DBP and CHL in each step. Note: Five marks (5 marks) will be allocated for successful combination of the results.
Question 6: Using appropriate statistical method(s) investigate whether patients who had a history of being treated with Drug-X on a regular basis have significantly different systolic blood pressure (SBP), diastolic blood pressure (DBP) and cholesterol level (CHL) compared to patients who were not treated with Drug-X. In this task you will need to:
Please fill out the following table from the above analysis results
Variable
|
Summary Statistics
|
p-value
|
95% CI (where appropriate)
|
Yes Drug-X
|
No Drug-X
|
Systolic blood pressure
|
|
|
|
|
Diastolic blood pressure
|
|
|
|
|
Cholesterol level
|
|
|
|
|
Question 7 Using an appropriate statistical method evaluate the significance of difference of SBP across the following categories of BMI: BMI <25, BMI 25 to <30 and BMI ≥ 30kg/m2.