Stata 15 help for ltable

[ST] ltable -- Life tables for survival data

Syntax

ltable timevar [deadvar] [if] [in] [weight] [, options]

timevar specifies the time of failure or censoring. If deadvar is not specified, all values of timevar are interpreted as failure times. Observations with timevar equal to missing are ignored.

deadvar specifies how the time recorded in timevar is to be interpreted. Observations with deadvar equal to 0 are treated as censored and all other nonmissing values indicate that timevar should be interpreted as a failure time. Observations with deadvar equal to missing are ignored.

deadvar does not specify the number of failures. Specify frequency weights for aggregated data recording the number of failures.

options Description ------------------------------------------------------------------------- Main notable display graph only; suppress display of table graph present the table graphically, as well as in tabular form by(groupvar) produce separate tables (or graphs) for each value of groupvar test report chi-squared measure of differences between groups (2 tests) overlay overlay plots on the same graph survival display survival table; the default failure display cumulative failure table hazard display hazard table ci graph confidence interval level(#) set confidence level; default is level(95) noadjust suppress actuarial adjustment to the number at risk tvid(varname) subject ID variable to use with time-varying parameters intervals(w|numlist) time intervals in which data are to be aggregated for tables saving(filename[, replace]) save the life-table data to filename; use replace to overwrite existing filename

Plot plotopts(plot_options) affect rendition of the plotted line and plotted points plot#opts(plot_options) affect rendition of the #th plotted line and plotted points; available only with overlay

CI plot ciopts(rspike_options) affect rendition of the confidence intervals ci#opts(rspike_options) affect rendition of the #th confidence interval; available only with overlay

Add plots addplot(plot) add other plots to the generated graph

Y axis, X axis, Titles, Legend, Overall twoway_options any options other than by() documented in [G-3] twoway_options byopts(byopts) how subgraphs are combined, labeled, etc. -------------------------------------------------------------------------

plot_options Description ------------------------------------------------------------------------- connect_options change look of lines or connecting method marker_options change look of markers (color, size, etc.) -------------------------------------------------------------------------

fweights are allowed; see weight.

Menu

Statistics > Survival analysis > Summary statistics, tests, and tables > Life tables for survival data

Description

ltable displays and graphs life tables for individual-level or aggregate data and optionally presents the likelihood-ratio and log-rank tests for equivalence of groups. ltable also allows you to examine the empirical hazard function through aggregation.

Options

+------+ ----+ Main +-------------------------------------------------------------

notable suppresses displaying the table. This option is often used with graph.

graph requests that the table be presented graphically, as well as in tabular form; when notable is also specified, only the graph is presented. When you specify graph, only one table can be calculated and graphed at a time; see survival, failure, and hazard below.

graph may not be specified with hazard. Use sts graph to graph estimates of the hazard function.

by(groupvar) creates separate tables (or graphs within the same image) for each value of groupvar. groupvar may be string or numeric.

test presents two chi-squared measures of the differences between groups, the likelihood-ratio test of homogeneity and the log-rank test for equality of survivor functions. The two groups are identified by the by() option, so by() must also be specified.

overlay causes the plot from each group identified in the by() option to be overlaid on the same graph. The default is to generate a separate graph (within the same image) for each group. This option requires the by() option.

survival, failure, and hazard indicate the table to be displayed. If none is specified, the default is the survival table. Specifying failure displays the cumulative failure table. Specifying survival failure would display both the survival and the cumulative failure table. If graph is specified, multiple tables may not be requested.

ci graphs the confidence intervals around survival, failure, or hazard.

level(#) specifies the confidence level, as a percentage, for confidence intervals. The default is level(95) or as set by set level.

noadjust suppresses the actuarial adjustment to the number at risk. The default is to consider the adjusted number at risk for each interval as the total at the start minus (the number of censored)/2. If noadjust is specified, the number at risk is simply the total at the start, corresponding to the standard Kaplan-Meier assumption. noadjust should be specified when using ltable to list results corresponding to those produced by sts list.

tvid(varname) is for use with longitudinal data with time-varying parameters. Each subject appears in the data more than once, and equal values of varname identify observations referring to the same subject. When tvid() is specified, only the last observation on each subject is used in making the table. The order of the data does not matter, and last here means the last observation chronologically.

intervals(w|numlist) specifies the intervals into which the data are to be aggregated for tabular presentation. A numeric argument is interpreted as the width of the interval. For instance, interval(2) aggregates data into the intervals 0<=t<2, 2<=t<4, and so on. Not specifying interval() is equivalent to specifying interval(1). Because in most data, failure times are recorded as integers, this amounts to no aggregation except that implied by the recording of the time variable, and so it produces Kaplan-Meier product-limit estimates of the survival curve (with an actuarial adjustment; see the noadjust option above). Also see [ST] sts list. Although it is possible to examine survival and failure without aggregation, some form of aggregation is almost always required for examining the hazard.

When more than one argument is specified, intervals are aggregated as specified. For instance, interval(0,2,8,16) aggregates data into the intervals [0,2), [2,8), [8,16), and (if necessary) [16, infinity).

interval(w) is equivalent to interval(0,7,15,30,60,90,180,360,540,720), corresponding to 1 week, (roughly) 2 weeks, 1 month, 2 months, 3 months, 6 months, 1 year, 1.5 years, and 2 years when failure times are recorded in days. The w suggests widening intervals.

saving(filename[, replace]) creates a Stata data file (.dta file) containing the life table. This option will not save the graph to disk; see [G-2] graph save to save the resulting graph to disk.

replace specifies that filename be overwritten if it exists. This option is not shown in the dialog box.

+------+ ----+ Plot +-------------------------------------------------------------

plotopts(plot_options) affects the rendition of the plotted line and plotted points; see [G-3] connect_options and [G-3] marker_options.

plot#opts(plot_options) affects the rendition of the #th plotted line and plotted points; see [G-3] connect_options and [G-3] marker_options. This option is valid only if overlay is specified.

+---------+ ----+ CI plot +----------------------------------------------------------

ciopts(rspike_options) affects the rendition of the confidence intervals for the graphed survival, failure, or hazard; see [G-3] rspike_options.

ci#opts(rspike_options) affects the rendition of the #th confidence interval for the graphed survival, failure, or hazard; see [G-3] rspike_options. This option is valid only if overlay is specified.

+-----------+ ----+ Add plots +--------------------------------------------------------

addplot(plot) provides a way to add other plots to the generated graph; see [G-3] addplot_option.

+-----------------------------------------+ ----+ Y axis, X axis, Titles, Legend, Overall +--------------------------

twoway_options are any of the options documented in [G-3] twoway_options, excluding by(). These include options for titling the graph (see [G-3] title_options) and for saving the graph to disk (see [G-3] saving_option).

byopts(byopts) affects the appearance of the combined graph when by() is specified, including the overall graph title and the organization of subgraphs. See [G-3] by_option.

Examples

--------------------------------------------------------------------------- Setup . webuse rat

List some of the data . list in 1/5 . list if died == 0

Display the life table for group 1 . ltable t died if group == 1

Display the life table for group 1 aggregated into 30-day intervals . ltable t died if group == 1, interval(30)

Display the life table for group 1 using the specified intervals . ltable t died if group == 1, interval(120,180,210,240,330)

Display separate life tables for each group and aggregate into 30-day intervals . ltable t died, by(group) interval(30)

Display a failure table for group 1 aggregated into 30-day intervals . ltable t died if group == 1, interval(30) failure

--------------------------------------------------------------------------- Setup . webuse selvin

List some of the data . list in 1/5

Obtain survival rates, and show both table and graph . ltable t died [freq=pop], graph

--------------------------------------------------------------------------- Setup . webuse tumor

List some of the data . list in 1/6, sep(0)

Display the hazard table . ltable t d [freq=pop], hazard interval(0(1)9) ---------------------------------------------------------------------------

Video example

How to construct life tables


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index