ADNI data is made available to researchers around the world. As such, there are many active research projects accessing and applying the shared ADNI data. To further encourage Alzheimer’s disease research collaboration, and to help prevent duplicate efforts, the list below shows the specific research focus of the active ADNI investigations. This information is requested annually as a requirement for data access.
| Principal Investigator | |
| Principal Investigator's Name: | Adam Gerstenecker |
| Institution: | University of Louisville |
| Department: | Psychological and Brain Sciences |
| Country: | |
| Proposed Analysis: | Statistical analyses will be conducted using SPSS version 20. To test hypotheses related to the screening model (see hypotheses section), receiving operating characteristic plots (ROC) will be calculated for each variable of interest. Using ROC plots, sensitivity and specificity for each variable can be obtained by taking sensitivity against 1-specificity (see Altman & Bland, 1994 for a detailed description of ROC plots). To test hypotheses related to individual predictors, comparison of predictor categories (i.e., neuropsychological data, biomarkers, etc.), and the conversion model, a series of Cox proportion hazards regression analyses will be conducted. Cox proportional hazards are commonly used to relate the effects of covariates to survival time. For the proposed study, dementia will be used in place of survival. In contrast to linear regression, Cox proportional hazards do not adhere to the assumptions of normality. For the first step, all variables of interest (see hypotheses section) will be analyzed via a Cox proportional hazards regression. These values will be reported to note the predictive power of individual variables and predictor categories. Next, factors demonstrating modest levels of predictive power (p≤.1) will be analyzed via a second Cox proportional hazards regression analysis. Using these variables, beta weights will be used as a risk score for each individual factor. Finally, each risk score will be multiplied by a constant so theycan be summed to form easy to interpret groups (i.e., 0-5=Low Risk for Conversion group, 6-10=Moderate Risk for Conversion group, and 11-15=High Risk for Conversion group). This process will also be completed adjusting for age, sex, education, and ethnicity. |
| Additional Investigators | |
| Investigator's Name: | Benjamin Mast |
| Proposed Analysis: | Statistical analyses will be conducted using SPSS version 20. To test hypotheses related to the screening model (see hypotheses section), receiving operating characteristic plots (ROC) will be calculated for each variable of interest. Using ROC plots, sensitivity and specificity for each variable can be obtained by taking sensitivity against 1-specificity (see Altman & Bland, 1994 for a detailed description of ROC plots). To test hypotheses related to individual predictors, comparison of predictor categories (i.e., neuropsychological data, biomarkers, etc.), and the conversion model, a series of Cox proportion hazards regression analyses will be conducted. Cox proportional hazards are commonly used to relate the effects of covariates to survival time. For the proposed study, dementia will be used in place of survival. In contrast to linear regression, Cox proportional hazards do not adhere to the assumptions of normality. For the first step, all variables of interest (see hypotheses section) will be analyzed via a Cox proportional hazards regression. These values will be reported to note the predictive power of individual variables and predictor categories. Next, factors demonstrating modest levels of predictive power (p≤.1) will be analyzed via a second Cox proportional hazards regression analysis. Using these variables, beta weights will be used as a risk score for each individual factor. Finally, each risk score will be multiplied by a constant so theycan be summed to form easy to interpret groups (i.e., 0-5=Low Risk for Conversion group, 6-10=Moderate Risk for Conversion group, and 11-15=High Risk for Conversion group). This process will also be completed adjusting for age, sex, education, and ethnicity. |