is each sibling's death pregnancy-related, by MICS coding?

MICS asks three separate binaries where the DHS uses one coded item: MM22 pregnant when she died, MM23 died during childbirth, and MM24 died within two months of the end of a pregnancy (MM10, MM11, MM12 in MICS4/5). Any of the three makes the death pregnancy-related, with no cause exclusion.

Usage

is_preg_related_mics(sib_df, preg.window = c("2months", "42days"))

Arguments

sib_df

the prepped sibling dataset

preg.window

width of the postpartum window: "2months" (default) or "42days". See Details

Value

a logical vector, one entry per row of sib_df

Details

preg.window chooses how wide the postpartum window is:

• "2months" (the default) takes MM24 = 1 at face value, so the window is however long the respondent understood "two months" to be. This is the widest reading and the one this package has always used for MICS.

• "42days" additionally requires MM25 < 42. This is the WHO definition of a pregnancy-related death, it is what UNICEF's own tabulation syntax computes, and it is therefore what published MICS tables report. It is also consistent with is_preg_related_dhs, which already applies a 42-day cut through the mm12 band 100–141.

Only "42days" reproduces published MICS figures. On Iraq 2018 it gives 64.4 pregnancy-related deaths against a published 64, and on Madagascar 2018 136.6 against a published 137; "2months" gives 67.7 and 140.4.