What are the experiences of MCHP researchers in longitudinal data analysis – using only the Repository, or a hybrid?
Much of what MCHP has labeled “longitudinal” research has been time series analyses. In other words, trends in something like hospitalization rates by year, or mortality rates over time, have given a sense of “longitudinal” analysis. This is different than following a cohort of people over time, in a prospective analysis (or retrospective analysis) of long-term outcomes (or risk factors). One report (Brownell et al. 2003) looked at the health status of Manitobans over time, to explore the issue of the widening gap between the most and least healthy. From 1985 through to 1998, it was discovered that the “gap” resulted not in worsening health status of the least healthy, but increasing healthiness of the most healthy (while the least healthy stayed the same). Another report on seniors (Menec et al. 2002) looked at trends in seniors dying from heart disease from 1984 through to 1999, concluding that mortality rates for heart disease declined significantly for both men and women for all ag
