PROJECT ABSTRACT The search for effective strategies to prevent or treat Alzheimer's disease and related causes of dementia (ADRD) is a global public health priority. This search has been exceptionally difficult, reflecting the complex, multi-factorial processes culminating in ADRD, the diagnostic ambiguity, and the intrinsic challenges of research on diseases of the brain. Methods in Longitudinal Research on Dementia (MELODEM) is an international initiative to strengthen, harmonize, and promulgate analytic and design approaches for ADRD research. Launched in 2012, MELODEM convenes an international interdisciplinary group of researchers, including epidemiologists, neuropsychologists, neurologists, and biostatisticians, across career stages, to identify and address common methodologic challenges in quantitative research on ADRD. These methodologic challenges—such as selection/attrition, reverse causation, and measurement of cognition and dementia—have the potential to bias estimates of the extent to which risk factors protect against ADRD or slow the progression of ADRD. Some biases can be so extreme as to reverse the direction of estimated effects (e.g., protective instead of harmful association). For the past 5 years, MELODEM has convened an annual in-person meeting of 30-40 researchers and monthly web-based conference calls of methodologically themed working groups. The proposed meeting grant will support annual in-person MELODEM meetings to foster development and dissemination of valid and rigorous analytical approaches for quantitative research on the prevention and treatment of ADRD, with the goals of (a) developing new methods when needed, (b) reaching consensus on “best practices” for analytic methods, and (c) providing a platform to share and teach methods to junior and senior researchers. To continue working towards rigorous analytical approaches and “best practices,” the MELODEM annual progress meeting will include both “Data Workshops” and “Discussion Sessions” on topics relevant to current methodological challenges in ADRD research. The sessions will be led and moderated by leaders in the field and researchers ranging from graduate students to senior scientists will participate. The grant will also provide travel support for US researchers to participate in the annual meeting and ancillary activities. Finally, to sustain the collaborations and ideas discussed during the meeting, the grant will support post-meeting activities culminating in published reports from the collaborative research stemming from the conference and connections with a broader group of researchers not able to attend the meetings in person. These activities include web-based conference calls and small working group meetings. Support for the proposed MELODEM meetings will strengthen quantitative research methods related to ADRD, foster a collaborative, interdisciplinary community addressing the most important technical challenges, and promote rapid adoption of best-practices for solving shared technical challenges in ADRD research on prevention and treatment of ADRD.

PROJECT NARRATIVE Methods in Longitudinal Dementia Research (MELODEM) is an international initiative that provides a forum for identifying and sharing effective methods for quantitative research on prevention or treatment of Alzheimer's disease and related causes of dementia (ADRD) among humans. MELODEM enlists researchers from a variety of disciplines to develop research methods; fosters interdisciplinary conversations and collaborations to understand which methods are most applicable; and accelerates the exchange of new methods between researchers. This grant would support five years of MELODEM annual meetings and associated activities to foster stronger analytic methods on ADRD.