Stata 15 help for dstdize

[R] dstdize -- Direct and indirect standardization

Syntax

Direct standardization

dstdize charvar popvar stratavars [if] [in], by(groupvars) [dstdize_options]

Indirect standardization

istdize casevar_s popvar_s stratavars [if] [in] using filename, {popvars(casevar_p popvar_p) | rate(ratevar_p {#|crudevar_p})} [istdize_options]

charvar is the characteristic to be standardized across different subpopulations identified by groupvars.

popvar defines the weights used in standardization.

stratavars defines the strata across which the weights are to be averaged in dstdize. For istdize, stratavars defines the strata for which casevar_s is measured.

casevar_s is the variable name for the study population's number of cases. If by(groupvars) is specified, casevar_s must be constant or missing within each group defined by combinations of groupvars.

popvar_s identifies the number of subjects in each strata in the study population.

filename must be a Stata dataset and contain popvar and stratavars.

dstdize_options Description ------------------------------------------------------------------------- Main * by(groupvars) study populations using(filename) use standard population from Stata dataset base(#|string) use standard population from a value of grouping variable level(#) set confidence level; default is level(95)

Options saving(filename) save computed standard population distribution as a Stata dataset format(%fmt) final summary table display format; default is %10.0g print include table summary of standard population in output nores suppress storing results in r() ------------------------------------------------------------------------- * by(groupvars) is required.

istdize_options Description ------------------------------------------------------------------------- Main * popvars(casevar_p popvar_p) for standard population, casevar_p is number of cases and popvar_p is number of individuals * rate(ratevar_p {#|crudevar_p}) ratevar_p is stratum-specific rates and # or crudevar_p is the crude case rate value or variable level(#) set confidence level; default is level(95)

Options by(groupvars) variables identifying study populations format(%fmt) final summary table display format; default is %10.0g print include table summary of standard population in output ------------------------------------------------------------------------- * Either popvars(casevar_p popvar_p) or rate(ratevar_p {#|crudevar_p}) must be specified.

Menu

dstdize

Statistics > Epidemiology and related > Other > Direct standardization

istdize

Statistics > Epidemiology and related > Other > Indirect standardization

Description

dstdize produces standardized rates, a weighted average of the stratum-specific rates.

istdize produces indirectly standardized rates that are appropriate when the stratum-specific rates for the population being studied are either unavailable or unreliable.

istdize also calculates a point estimate and exact confidence interval for the study population's standardized mortality ratio (SMR) or the standardized incidence ratio.

Options for dstdize

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

by(groupvars) is required for the dstdize command; it specifies the variables identifying the study populations. If base() is also specified, there must be only one variable in the by() group. If you do not have a variable for this option, you can generate one by using something like generate newvar=1 and then use newvar as the argument to this option.

using(filename) or base(#|string) may be used to specify the standard population. You may not specify both options. using(filename) supplies the name of a .dta file containing the standard population. The standard population must contain the popvar and the stratavars. If using() is not specified, the standard population distribution will be obtained from the data. base(#|string) lets you specify one of the values of groupvar -- either a numeric value or a string -- to be used as the standard population. If neither base() nor using() is specified, the entire dataset is used to determine an estimate of the standard population.

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

+---------+ ----+ Options +----------------------------------------------------------

saving(filename) saves the computed standard population distribution as a Stata dataset that can be used in further analyses.

format(%fmt) specifies the format in which to display the final summary table. The default is %10.0g.

print includes a table summary of the standard population before displaying the study population results.

nores suppresses storing results in r(). This option is seldom specified. Some results are stored in matrices. If there are more groups than matsize, dstdize will report "matsize too small". Then you can either increase matsize or specify nores. The nores option does not change how results are calculated but specifies that results need not be left behind for use by other programs.

Options for istdize

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

popvars(casevar_p popvar_p) or rate(ratevar_p {#|crudevar_p}) must be specified with istdize. Only one of these two options is allowed. These options are used to describe the standard population's data.

With popvars(casevar_p popvar_p), casevar_p records the number of cases (deaths) for each stratum in the standard population, and popvar_p records the total number of individuals in each stratum (individuals at risk).

With rate(ratevar_p {#|crudevar_p}), ratevar_p contains the stratum-specific rates. #|crudevar_p specifies the crude case rate either by a variable name or by the crude case rate value. If a crude rate variable is used, it must be the same for all observations, although it could be missing for some.

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

+---------+ ----+ Options +----------------------------------------------------------

by(groupvars) specifies variables identifying study populations when more than one exists in the data. If this option is not specified, the entire study population is treated as one group.

format(%fmt) specifies the format in which to display the final summary table. The default is %10.0g.

print outputs a table summary of the standard population before displaying the study population results.

Examples

--------------------------------------------------------------------------- Setup . webuse hbp . generate pop = 1

Obtain standardized rates of hbp by city and year, using the age, race, and sex distribution of the cities and years combined as the standard . dstdize hbp pop age race sex, by(city year)

--------------------------------------------------------------------------- Setup . webuse kahn, clear

Obtain mortality rates by state using the standard population saved in popkahn.dta . istdize death pop age using http://www.stata-press.com/data/r15/popkahn, by(state) pop(deaths pop) print ---------------------------------------------------------------------------

Stored results

dstdize stores the following in r():

Scalars r(k) number of populations

Macros r(by) variable names specified in by() r(c#) values of r(by) for #th group

Matrices r(se) 1 x k vector of standard errors of adjusted rates r(ub_adj) 1 x k vector of upper bounds of confidence intervals for adjusted rates r(lb_adj) 1 x k vector of lower bounds of confidence intervals for adjusted rates r(Nobs) 1 x k vector of number of observations r(crude) 1 x k vector of crude rates (*) r(adj) 1 x k vector of adjusted rates (*) (*) If, in a group, the number of observations is 0, then 9 is stored for the corresponding crude and adjusted rates.

istdize stores the following in r():

Scalars r(k) number of populations

Macros r(by) variable names specified in by() r(c#) values of r(by) for #th group

Matrices r(cases_obs) 1 x k vector of number of observed cases r(cases_exp) 1 x k vector of number of expected cases r(ub_adj) 1 x k vector of upper bounds of confidence intervals for adjusted rates r(lb_adj) 1 x k vector of lower bounds of confidence intervals for adjusted rates r(crude) 1 x k vector of crude rates r(adj) 1 x k vector of adjusted rates r(smr) 1 x k vector of SMRs r(ub_smr) 1 x k vector of upper bounds of confidence intervals for SMRs r(lb_smr) 1 x k vector of lower bounds of confidence intervals for SMRs


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