STATISTICAL ANALYSIS
To investigate the relationships between clinical and demographic variables with
Parkinson's Disease severity and onset, we will employ the following hypothesis
tests:
T-tests:
Objective:
To compare the means of continuous clinical or cognitive performance measures
(e.g., UPDRS score, MoCA score) between diagnosed and non-diagnosed groups.
Hypothesis:
Null Hypothesis (H0H_0H0): There is no significant difference in cognitive
performance (MoCA score) between the two groups (diagnosed and non-diagnosed).
Alternative Hypothesis (HAH_AHA): There is a significant difference in cognitive
performance between the two groups.
Chi-square Tests:
Objective: To assess associations between categorical variables (e.g., smoking
status, gender, family history) and diagnosis of Parkinson's Disease.
Hypothesis:
Null Hypothesis (H0H_0H0): There is no association between family history and
Parkinson's Disease diagnosis.
Alternative Hypothesis (HAH_AHA): There is a significant association between family
history and Parkinson's Disease diagnosis.
ANOVA:
Objective: To compare the means of a continuous outcome (e.g., UPDRS or MoCA
scores) across multiple groups based on disease severity.
Hypothesis:
Null Hypothesis (H0H_0H0): There is no significant difference in UPDRS scores
across the different severity levels of Parkinson’s Disease.
Alternative Hypothesis (HAH_AHA): There is a significant difference in UPDRS scores
across the different severity levels.
Regression Analysis (Linear and Logistic Regression)
Multiple Linear Regression:
To explore how clinical, lifestyle, and demographic variables relate to the severity
of Parkinson’s Disease (measured by a continuous outcome like UPDRS score), I
will use multiple linear regression.
Objective: To quantify the relationship between Parkinson’s Disease severity
(UPDRS score) and predictor variables such as age, BMI, physical activity, alcohol
consumption, and cognitive performance (MoCA score).
Expected Outcome: This analysis will reveal which factors are most strongly
associated with higher UPDRS scores, indicating greater disease severity.
Logistic Regression:
To assess the factors associated with the likelihood of being diagnosed with
Parkinson's Disease, I will employ logistic regression analysis.
Objective: To estimate the odds of Parkinson’s Disease diagnosis based on
variables like age, smoking status, family history of Parkinson’s, and cognitive
performance (MoCA score).
Expected Outcome: The logistic regression model will identify factors that
increase the likelihood of a Parkinson’s Disease diagnosis, offering insights into risk
factors and possible early intervention targets.