Following is a response from Senior DHS staff member, Tom Pullum:
The following Stata lines show how to construct the indicators for the columns of this table. You may not have realized that this is an example of multiple options: it is possible for a respondent to have more than one type of insurance. My n is low by 4 cases--for reasons unknown.
use " KEIR72FL.DTA", clear
tab1 v481*,m
* Sequence as in table 14.10.1
gen ins1=100*v481c if v481<.
gen ins2=100*v481b if v481<.
gen ins3=100*v481a if v481<.
gen ins4=100*v481d if v481<.
gen ins5=100*v481e if v481<.
gen ins6=100*v481x if v481<.
label variable ins1 "National"
label variable ins2 "Employer"
label variable ins3 "Mutual"
label variable ins4 "Private"
label variable ins5 "Pre-payment"
label variable ins6 "Other"
gen none=100
replace none=0 if v481==0
tabstat ins* none [fweight=v005], statistics(mean)
tab v025 if v481<. [iweight=v005/1000000]