Problem with add_covariate_to_timeline

[krishnaaswin77]

I am performing Cox regression with the time-varying covariates method on a dataset.

I use to_long_format from lifelines.utils to convert the data into a long format. There are 7092 events (rows) in the data in the long format.

For some events, the value of a variable X changes over time, and I include this in a type 1 format table as described in the documentation: https://lifelines.readthedocs.io/en/latest/Time%20varying%20survival%20regression.html#dataset-creation-for-time-varying-regression
These correspond to 8936 rows / changes.

Now I combine them using the add_covariate_to_timeline function, which used to work before. But now for some reason, the output dataframe with the two tables folded into each other give out a 7092 row-sized table. The output table is not taking into account the changes in X over time (all start values in the output table are 0).

I am not sure why add_covariate_to_timeline is not working, any leads would be appreciated.