Table 11.3 HERS data

Description

Dataset used in Chapter 11 of Regression Methods in Biostatistics (2nd edition), as distributed on the UCSF companion website. Study design: Baseline/Visit 1 subset from the longitudinal HERS data structure. Primary article: Hulley and colleagues (1998), JAMA.

Usage

hers_long_base_visit1_only_saved

Format

A data frame with 5526 rows and 22 variables:

pptid
Participant identifier.
nvisit
Visit number (0=baseline)
age
Age or age category at baseline (as coded in source data).
visit
Visit number (time point) in longitudinal follow-up.
exer3
bl exercise &/or walk >3x/wk, 1:y 0:n
avgdrpwk
avg drinks per week
csmker
current smoking status 1:current
weight
Body weight (kg).
bmi
body mass index
sbp
average systolic blood pressure
glucose
Fasting plasma glucose (mg/dL).
htnmeds
anti-htn meds use
drinkany
(1 if avgdrpwk>0, 0 otherwise)
diabetes
(1=yes, 0=no)
highsbp
Indicator for elevated systolic blood pressure.
sbp_z
Standardized (z-score) systolic blood pressure.
group
treatment assignment 0:a: pbo, 1:b:hrt
grade_hi
highest grade completed
white
(1=white, 0=other)
baseline_dm
Indicator for diabetes at baseline.
miss_mar
Indicator for values set missing under MAR mechanism.
miss_nmar
Indicator for values set missing under NMAR mechanism.

Source

UCSF companion data