For fertility, we take the average total fertility rate (F) for the most recent five-year interval preceding the H-NTA survey date, using United Nations quinquennial data.
The analysis sample of 1,606 includes all women who had a birth three years prior to the survey date and who had no missing data.
Since deaths during the recall period >1 to ≤5 years prior to the survey date only captures information on children up to age four we calculated overall and cause-specific neonatal, child (1 48 months), and under-4 mortality rates separately using recall data from ≤1 year prior to the survey date and from >1 to ≤5 years prior to the survey date (4 years duration).
Remaining data were collected only at the time closest to the survey date (lipid levels, serum creatinine, estimated glomerular filtration rate, height, weight, and blood pressure).
We also used the ART register to identify the newly treated cohort of patients who started ART during the period from 12 months prior to the baseline survey date up to 3 months before the date of final data collection.
HbA1c and fasting blood glucose data were collected at the time closest to diagnosis and time closest to the survey date (i.e., most recent laboratory measurements).
Patients present on the survey date were included.
4. List the rent survey date (or range of dates).
The effects are already visible a few years prior to the date of data collection, and our fertility measures overestimate the true first-birth fertility by some ten percent at 15 20 years prior to the simulated survey date.
The samples and data that were used had been collected for a different purpose in a population survey dating back to 1981 [ 12].
These factors -- known collectively as the "house effect" -- include survey dates, question wording and order, sampling and calling procedures, and the way pollsters weight their data.
