Table 1.
Description of scenario analyses to test assumptions related to missing data and loss-to-follow-up.
Table 2.
Characteristics of community or non-nursing home residential care dwelling older adults at the time of the Round 1 National Health and Aging Trends Study interview a.
Table 3.
Reasons for loss-to-follow-up, stratified by baseline frailty phenotype a.
Fig 1.
Sankey diagram of frailty state transitions using inverse probability of censoring weighting to address potential informative censoring.
Table 4.
Results of scenario analyses to test assumptions related to missing data and loss-to-follow-up.