lecture 10 for BioEpi 691F

Practical Data Management and Statistical Computing (BioEp691F)

Outline

Assignments

Outline: Lec11 Lec12 Lec13 lec14 Lec15 Lec16 Lec17 Lec18 Lec19 Lec20
Lectures: Lec11 Lec12 Lec13 Lec14 Lec15 Lec16 Lec17 Lec18 Lec19 Lec20

Lecture 11

1. Reading Data With Multiple and Unequal Numbers of Lines per Subject (An Example)

Chronic Granulomatous Disease (CGD) is a group of inherited rare disorders of the immune function characterized by recurrent pyogenic infections which usually present early in life and may lead to death in childhood. There is evidence establishing a role for gamma interferon as an important macrophage activating factor which could restore superoxide anion production and bacterial killing by phagocytes in CGD patients. In order to study the ability of gamma interferon to reduce the rate of serious infections, that is, the rate of infections requiring hospitalization for parenteral antibiotics, a double-blinded clinical trial was conducted in which patients were randomized to placebo vs. gamma interferon. The UMASS data set contain the data and a brief description. The research hypothesis is:

Does the infection rate differ between the two intervention groups?

Analysis Plan: Create infection rate variable for each subject over time. Separate cross-sectional from longitudinal data. Summarize infection rates for longitudinal data. Compare rates between treatment groups overall, and by gender.

Inspection of the data indicates that there are multiple infections per subject. We read the data for a subset of the variables ( ID, Z1,Z8,T1, and D) in dmes99p19.sas. The resulting data set has 1 record for each infection. We focus on the variables ID, IDT , T1 and D.

We separate variables into a cross-sectional, and a longitudinal data set in dmes99p20.sas.

creating multiple data sets in a DATA step.
- OUTPUT statement
- Using the automatic variable FIRST.xxx with BY xxx; statment when Data has been SORTED ; BY xxx;

Add a record to the longitudinal data set for the patient entry time, and recode the events to 0, 1 (dmes99p21.sas).

use SET to concatenate two data sets.

Create variable that represents the cumulative number of infections for each subject (dmes99p22.sas)

Using the automatic variable FIRST.xxx with BY xxx; statment when Data has been SORTED ; BY xxx;
Use RETAIN statement in DATA step to cumulate infection for each subject.

Add cross-sectional variables to longitudinal data (dmes99p23.sas)

use MERGE statement in DATA step with BY xxx; statment when Data has been SORTED ; BY xxx;

Plot simultaneously Number of Infections over time by Treatment. (dmes99p24.sas)

use PROC GPLOT with options
- SYMBOL statement options
  - I=JOIN to connect points
  - C=BLACK to select color
  - R=100 to repeat specification for subsequent SYMBOL# statements
- BY c to create separate plots for levels of c
- PLOT a*b=c to overlay separate plots for levels of c
  - option NOLEGEND to suppress the legend for subjects.

Produced and maintained by the Dept of BioEpi at UMASS
Send comments or questions about this web site to Ed Stanek
Email: stanek@schoolph.umass.edu
\be691f\web\webready\lec11.html
Lst Update: 10/21/99