Longitudinal Family Analysis of Blood Pressure
Author Information
Author(s): Julia MP Soler, John Blangero
Primary Institution: Department of Statistics, University of Sao Paulo, Brazil; Department of Genetics, Southwest Foundation for Biomedical Research, USA
Hypothesis
Can a log-linear model effectively analyze longitudinal familial data for blood pressure?
Conclusion
The study suggests that corrections for censored blood pressure data can retain original variability and that empirical semi-variogram plots are useful for diagnosing variance models.
Supporting Evidence
- The study analyzed systolic blood pressure data from 1667 individuals across multiple measurements.
- Empirical semi-variogram plots indicated that the blood pressure dispersion pattern followed the assumptions of the model.
- The methodology allows for corrections of censored data while retaining original variability.
Takeaway
This study looks at how family history affects blood pressure over time and finds ways to correct data when some measurements are missing.
Methodology
The study used a mixed-model longitudinal approach to analyze systolic blood pressure data from family members, correcting for right-censored values due to hypertension treatment.
Potential Biases
Potential bias due to the reliance on self-reported hypertension treatment and the handling of censored data.
Limitations
The study primarily focuses on systolic blood pressure and may not generalize to other traits.
Participant Demographics
Participants included 822 males and 845 females from two cohorts.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website