Measuring Coverage in MNCH: Total Survey Error and the Interpretation of Intervention Coverage Estimates from Household Surveys

Thomas P Eisele,Joseph Keating,Dale A Rhoda,Aluisio J D Barros,Felicity T Cutts,Fred Arnold,Ruilin Ren,Nyovani Madise

doi:10.1371/journal.pmed.1001386

Thomas P Eisele, Joseph Keating + Show 6 more

Open Access

https://doi.org/10.1371/journal.pmed.1001386

Copy DOI

Abstract

Nationally representative household surveys are increasingly relied upon to measure maternal, newborn, and child health (MNCH) intervention coverage at the population level in low- and middle-income countries. Surveys are the best tool we have for this purpose and are central to national and global decision making. However, all survey point estimates have a certain level of error (total survey error) comprising sampling and non-sampling error, both of which must be considered when interpreting survey results for decision making. In this review, we discuss the importance of considering these errors when interpreting MNCH intervention coverage estimates derived from household surveys, using relevant examples from national surveys to provide context. Sampling error is usually thought of as the precision of a point estimate and is represented by 95% confidence intervals, which are measurable. Confidence intervals can inform judgments about whether estimated parameters are likely to be different from the real value of a parameter. We recommend, therefore, that confidence intervals for key coverage indicators should always be provided in survey reports. By contrast, the direction and magnitude of non-sampling error is almost always unmeasurable, and therefore unknown. Information error and bias are the most common sources of non-sampling error in household survey estimates and we recommend that they should always be carefully considered when interpreting MNCH intervention coverage based on survey data. Overall, we recommend that future research on measuring MNCH intervention coverage should focus on refining and improving survey-based coverage estimates to develop a better understanding of how results should be interpreted and used.

Highlights

Representative household surveys are increasingly relied upon to measure maternal, newborn, and child health (MNCH) intervention coverage at the population level in low- and middle-income countries
Survey/s (MICS), the AIDS Indicator Survey/s (AIS), and the Malaria Indicator Survey/s (MIS). These surveys rely on scientific sampling methods, which require each element of the target population to have a known and non-zero probability of selection, to obtain point estimates of MNCH intervention coverage at the national and sub-national levels every 3–5 years [1]
In the absence of a formal hypothesis test, informal conclusions about the statistical significance of a difference are sometimes drawn by examining whether two confidence intervals overlap

Summary

Introduction

Representative household surveys are increasingly relied upon to measure maternal, newborn, and child health (MNCH) intervention coverage at the population level in low- and middle-income countries. In this review, which is part of the PLOS Medicine ‘‘Measuring Coverage in MNCH’’ Collection, we discuss the important sampling error and non-sampling error issues involved in interpreting MNCH intervention coverage estimates derived from household surveys for decision making using relevant examples from national surveys to provide context.

Results

Conclusion