SAS Project
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
SAS Project
Your poster may include (but is not limited to) the following sections:
1. Abstract: Summarize the entire study
2. Objective: Introduce the motivation and objective of the study
3. Methods: Provide a summary of the data and describe the statistical methodology.
4. Results: Report the main findings (descriptive statistics, model interpretations, etc.).
5. Discussion: Summarize the analysis and state its values and limitations.
6. Bibliography: If applicable, introduce some references.
Obesity: Bariatric Surgery
Obesity is one of the greatest public health problems in industrialized countries. According to data from the National Health Nutrition Examination Survey, more than 2 in 3 adults in the US are considered to be overweight or obese. Obesity is associated with an increased risk for type 2 diabetes, hypertension, dyslipidemia, cardiovascular disease, musculoskeletal disorders, certain types of cancers, and mortality.
In general, treatments for obesity can be classified into two categories: non-surgical therapy and bariatric surgery. The non-surgical therapy includes a variety of approaches such as behavioral therapy and dietary changes. The most commonly used bariatric surgery techniques are laparoscopic adjustable gastric banding (LAGB) and Roux-en-Y gastric bypass (RYGB). Current guidelines recommend the evaluation of bariatric surgery for individuals with a body mass index (BMI) >35 with serious comorbidities related to obesity.
With a primary hypothesis that surgical therapy for obesity achieves greater weight loss than non-surgical therapy, a group of researchers conducted a randomized clinical trial. Eligible patients (N = 450) were randomly allocated to receive a non- surgical therapy (control), LAGB (treatment 1), or RYGB (treatment 2). The patients’ height and weight were measured at baseline, 6, 12, 18, and 24 months after randomization and converted into BMI. Furthermore, at each visit, the patients were asked to complete the Compulsive Behaviors Questionnaire (CBQ), and the number of alcoholic drinks consumed per week was recorded.
The dataset [bmi.xlsx] contains data related to this study and contains variables defined as follows:
Variable |
Description |
ID |
Subject ID |
Gender |
Gender . . . { 0 = Male / 1 = Female } |
Diabetes |
Diabetes status at baseline . . . { 0 = No / 1 = Yes } |
Hypertension |
Hypertension status at baseline . . . { 0 = No / 1 = Yes } |
TRT |
Treatment . . . { 1 = Control / 2 = LAGB / 3 = RYGB } |
BMI0 – BMI4 |
BMI measured at baseline, 6, 12, 18, and 24 months |
DRINK0 – DRINK4 |
Number of drinks per week consumed at baseline – 24 months |
The researchers are interested in the association of weight loss with treatments. They are also concerned if gender, diabetic status, hypertension status, and substance abuse (alcohol consumption) are associated with the outcome of interest.
Some potential approaches include (but are not limited to):
1. Using the last measurement only and fitting a cross-sectional model
2. Incorporating the repeated measures and investigating the association over time
The BMI and DRINK measures were recorded as missing (NA) if lost to follow-up. As part of your descriptive data analysis, be sure to investigate the amount of missing data and pattern of missingness for these two variables.
After performing the appropriate analysis, interpret your final model and discuss the findings. Interaction terms can be included if considered reasonable.
Create descriptive statistics (tables + plots) for all relevant variables. You may not include every step of your hypothesis testing or model selection in the poster, but be sure to check all relevant assumptions for any hypothesis test performed or for any models built.
2023-04-28