Related Experiment Videos
An application of hierarchical linear models to longitudinal studies
1School of Nursing, State University of New York at Buffalo 14214, USA.
Research in Nursing & Health
|February 1, 1996
Summary
Hierarchical linear models (HLM) offer a flexible approach for nursing research on patient outcomes over time. HLM overcomes limitations of traditional methods, handling missing data and variable time intervals for precise patient status analysis.
Area of Science:
- Nursing Research
- Longitudinal Data Analysis
- Patient Outcomes
Background:
- Nursing research frequently examines patient outcome changes over time.
- Traditional methods (univariate/multivariate repeated measures, pre-post tests) have restrictive assumptions and data needs.
- A more flexible analytical approach is needed for complex longitudinal patient data.
Purpose of the Study:
- To introduce Hierarchical Linear Models (HLM) as a superior alternative for analyzing longitudinal patient outcomes in nursing research.
- To highlight the advantages of HLM over conventional statistical methods.
Main Methods:
- Application of Hierarchical Linear Models (HLM) for analyzing patient outcome data across time.
- Comparison of HLM with traditional repeated measures and pre-post test designs.
Main Results:
- HLM effectively describes individual growth trajectories and their relation to initial status.
- HLM accommodates missing data and does not require fixed time intervals.
- HLM provides more precise estimations compared to traditional methods.
Conclusions:
- Hierarchical Linear Models (HLM) present a flexible and robust method for nursing research on patient outcomes.
- HLM addresses limitations of traditional analyses, improving the study of patient changes over time.