CASE
STUDY ABSTRACT, OBJECTIVES & DESIGN
CASE
STUDY AMENDMENT: IMPACT OF 9/11 ON PWH/As TAKING HAART
CASE STUDY DATA ANALYSIS
PLAN
CASE
The Center for Adherence Support Evaluation
at the New York Academy of Medicine
CASE: Statistical Analysis Plan
Measuring Adherence Using Self Reports
- 3 day self report: Adherence rate = 1 - doses missed in last 3 days/prescribed
doses for 3 days
- last missed
- days per week missed at least one dose
- difficulty taking medication on time
- missed doses last weekend
- difficulty following special instructions
- reasons for missing doses in the last month
Testing Reliability and Validity of Self-Reports
- correlation analysis
- confirmatory factor analysis
Program Level Variables
- obtained from qualitative interviews of providers
- artificially generated program level variables using cluster analysis of service data
Effectiveness of Interventions in Supporting Adherence
- hierarchical linear models with repeated measures; modeling adherence
measure (transformed), biological markers and other improvement markers
with program level variables as predictors - change from baseline as
well as baseline as a covariate
- will also reflect any threshold time effects
- semiparametric modeling of adherence measure and biological markers
treating follow-up windows as panels
- a continuous approach
- generalized estimating equation modeling for categorical measures
Appropriate Levels of Service Utilization
- hierarchical linear models using services and tools as predictors
- categorical data analysis approach
Durability of the Intervention Effectiveness
- survival analysis approach to estimate median time to drop below specified levels of adherence after completion
Effectiveness in Supporting Specific Populations
- analyses controlling for client level variables
Success Through Achievement of Intermediate Objectives
- structural equation modeling
CASE
The New York Academy of Medicine
1216 Fifth Avenue
New York, NY 10029
(212) 822-7237
http://www.case.nyam.org
Last updated: 4/11/02