Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/414883
Title: | Estimation of Mean Transition Time using Markov Model and Comparison of risk factors of malnutrition using Markov Regression to Generalized Estimating Equations and Random Effects Model in a Longitudinal study |
Researcher: | Visalakshi J |
Guide(s): | |
Keywords: | Longitudinal study Malnutrition Markov Model Markov Regression Mean Transition Time Random Effects Model |
University: | The Tamil Nadu Dr. M.G.R. Medical University |
Completed Date: | 2012 |
Abstract: | Malnutrition refers to many diseases each with a specific deficiency in one or more nutrients and each characterized by cellular imbalance between the supply of nutrients and energy on the one hand, and the body s demand for them to ensure growth maintenance. Malnutrition is an important indicator of child health. It is now recognized that 6.6 million out of 12.2 million deaths among children under-five or 54% of young child mortality in developing countries is associated with malnutrition. India has the highest percentages of undernourished children in the world. During 1982, seven localities and 22 villages were selected for this study. These localities and villages were selected from Vellore town and KV Kuppam development block sampling frames respectively. All children aged 5-7 years were screened for signs of malnutrition by consultant pediatricians. The main aim was to find the risk factors for malnutrition. The objectives of the study are: (i) To estimate the first mean passage time which indicates the average time spent by a child to move from one state to another, to find risk factors of using GEE and Random Effects model, to find risk factors of protein energy malnutrition using transition probabilities, to find the risk factors by calculating the transition intensity matrices and to compare the results obtained from GEE and Markov regression models using transition probabilities and transition intensities. Conclusion: In any longitudinal study with discrete non-absorbing outcome, it is essential to estimate the duration of time spent in each state of the outcome. This will help us to study the impact of duration of stay with other risk factors. In longitudinal data if the current state of the outcome depends on the state of the outcome at the previous time, then Markov regression is the best approach to find the risk factors. GEE approach evaluates the overall correlation structure and therefore more likely to have larger standard errors and thereby likely to deal with false positive findings. |
Pagination: | 218 |
URI: | http://hdl.handle.net/10603/414883 |
Appears in Departments: | Department of Medical |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 27.03 kB | Adobe PDF | View/Open |
05_chapter 1.pdf | 62.98 kB | Adobe PDF | View/Open | |
06_chapter 2.pdf | 29.86 kB | Adobe PDF | View/Open | |
07_chapter 3.pdf | 233.33 kB | Adobe PDF | View/Open | |
08_chapter 4.pdf | 36.89 kB | Adobe PDF | View/Open | |
09_chapter 5.pdf | 265.29 kB | Adobe PDF | View/Open | |
10_annexures.pdf | 199.54 kB | Adobe PDF | View/Open | |
10_chapter 6.pdf | 443.91 kB | Adobe PDF | View/Open | |
11_chapter 7.pdf | 67.82 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 84.91 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: