Skip to main content

Table 4 Potential effect of missing routine PMTCT HIV status on estimated HIV prevalence in Mozambique

From: Routine data from prevention of mother-to-child transmission (PMTCT) HIV testing not yet ready for HIV surveillance in Mozambique: a retrospective analysis of matched test results

A. ANC surveillance HIV prevalence by routine PMTCT HIV test status

 

Total

ANC test results by availability of PMTCT rapid test

  

Due to refusal

Due to reasons other than refusal or known positive

  

PMTCT HIV status missing due to refusal

PMTCT HIV status not missing due to refusal

PMTCT HIV status missing due to other reasons

PMTCT HIV status not missing due to other reasons

HIV+

3283

9

3274

94

3189

Total

21222

83

21139

511

20711

HIV Prevalence

15.5%

10.8% (6.6-15.0)

15.5% (12.8-18.2)

18.4% (12.5-24.3)

15.4% (12.7-18.1)

B. Measures of effect of missing test results on HIV prevalence

Measure

Due to refusal

Due to reasons other than refusal or known positive

Absolute difference1

−4.6%

3.0%

Relative difference2

−30.0%

19.5%

Percent bias3

0.12%

−0.47%

Prevalence ratio4

0.70

1.19

  1. Notes:
  2. 1. Absolute difference is the HIV prevalence in those whose PMTCT HIV status is missing minus the HIV prevalence in those for whom it is present.
  3. 2. Relative difference is the absolute difference divided by the HIV prevalence among those for whom the PMTCT HIV status is present.
  4. 3. Percent bias is the HIV prevalence in those for whom PMTCT HIV status is present minus the overall HIV prevalence, divided by the overall HIV prevalence.
  5. 4. The prevalence ratio is the prevalence amongst those for whom the PMTCT HIV status is missing divided by the prevalence among those for whom the PMTCT HIV status is not missing.