## Info

1 Event status (D dead, A alive) and survival age 2 Event status (D dead, A alive) and age in 1975 (at start of observation period) Figure 1. Lifetimes of 12 hypothetical patients. Analytic approaches that rely on person-years at risk can underestimate hazards and incidence rates. For example, the hazard at age 70 years is the conditional probability of event onset at age 70, or the number of events at age 70 divided by the number of individuals at risk at age 70. Individuals who contribute...

## Data Preparation And Data Reduction

The data analysis for the nursing home institutionalization function consisted of two phases. First, there was the data preparation and reduction phase and second there was the development of the risk appraisal functions separately for each sex. The objective of the first phase was to insure the availability of stable, reliable variables for inclusion in the model. We now discuss the As noted already, there were a large number of variables or indicators, many measuring the same underlying...

## Software For Analysing Longitudinal Data

Unfortunately, most published methodology for analysing longitudinal data has not been incorporated into commercial software. For most of the methods discussed in this paper one needs to either develop new software or write to the authors for their research programs. Fortunately, there are some notable exceptions. Programs for fitting models to Gaussian data (that is, linear and non-linear mixed models and models with time-series error structure) and GEE for continuous and discrete data are...

## Extended Continuation Ratio Model

By comparing Figures 4 and 5 it is seen that the CR model in its ordinary form has no advantage over the PO model for this data set. The PO model has been extended by Peterson and Harrell15 to allow for unequal slopes for some or all of the X's for some or all levels of Y. This partial PO model requires specialized software, and with the demise of SAS Version 5 PROC LOGIST, software is not currently available. Armstrong and Sloan6 and Berridge and Whitehead8 showed how the CR model can be...

## Examples

Tions about when the events took place, for example at the beginning, midpoint or end of each The breast cancer data set Table I , described more fully in Finkelstein and Wolfe,8 consists of 94 observations from a retrospective study looking at the time to cosmetic deterioration. Information is available on one covariate, the type of therapy - either radiation alone coded 0 , or in combination with chemotherapy coded 1 . Of the 94 observations, 56 are interval-censored The AIDS data9 are...