Antenatal Coverage Looks Good on Paper: Verifying MCH Indicators
A national MCH programme reports 88% antenatal care (ANC) coverage (at least one visit) and 61% coverage of four or more visits, based on facility registers. A household survey in two districts (n=1,850 women with a birth in the past 2 years) finds 79% with at least one ANC visit and 34% with four or more visits. The survey also finds that only 46% of women had their first visit before 16 weeks, and that 28% of deliveries were attended by a skilled provider, against a register-based figure of 55%. The programme's maternal mortality ratio has not changed in 5 years. The ministry's proposed response is a media campaign urging women to 'attend ANC early and often'.
1. Reconcile the register and survey figures. What explains the gap, and what does it imply about the reliability of routine data?
Read the worked answer
Reconciliation
- The register counts visits, the survey counts women. A woman attending 6 times contributes 6 register entries but one woman; numerator inflation is structural, not fraudulent.
- Denominators differ: registers use facility catchment estimates, which over-count women who deliver or move elsewhere and under-count those who never attend.
- The skilled-birth-attendance gap (55% vs 28%) is too large for measurement noise alone and suggests register entries record 'attendance at a facility' rather than 'attended by a skilled provider', or count deliveries in facilities serving women from outside the catchment.
Implication
- The 88% figure cannot be used for planning or for accountability. The four-or-more figure (61% vs 34%) is the more decision-relevant indicator and is also the most inflated.
- The static maternal mortality ratio alongside 'high' ANC coverage is itself a signal: coverage that does not translate into outcomes is either mis-measured or not effective. Both must be examined.
Data governance: routine registers are useful for facility management but need periodic validation against a household survey or lot quality assurance sampling before being used as national indicators.
2. Evaluate the proposed media campaign and propose an alternative response grounded in the survey findings.
Read the worked answer
Why the campaign is likely to fail
- It targets demand, but the survey shows a timing problem (only 46% before 16 weeks) and a quality/access problem (28% skilled attendance). If women present early and are turned away, wait hours, or are treated poorly, a campaign increases frustration and erodes trust.
- It is untargeted: a mass message reaches the 79% who already attend and does little for the 21% who do not, who are likely the poorest and most remote.
- It does not address the register inflation, so the ministry will keep measuring the wrong thing.
Alternative response
- Diagnose the 21%: map who does not attend by wealth quintile, distance, and education; the survey data allow this.
- Fix supply before demand: staffing and stock at the facilities serving non-attenders; ensure a first visit can be completed in one trip (screening, supplements, and counselling together).
- Targeted demand-side measures: community health workers with lists of pregnant women; vouchers or transport support for the most remote; conditional incentives where evidence supports them.
- Quality of care: respectful maternity care training, and measure it (e.g., observed provider practice, client exit interviews).
- Measurement reform: adopt the survey-validated indicators; add a data-quality audit; track early ANC booking and skilled attendance as the primary indicators.
- Evaluate: define a comparison district and pre-specified indicators before launch.