Table 2

Study characteristics

Study, countryObjectiveParticipants
(dataset)
Variables (relevant for this review)Study methodsResultsResults summary
Abir et al (2014)
Bangladesh32
Identify factors associated with mortality in children under 5 years of ageSurvival information from Bangladesh DHS (2004–2011) of 16 722 singleton live-born infants of the most recent birth of mother within 3 years before the mother being interviewed
  1. Maternal marital status

  2. Religion

  3. Mother’s age

  4. Mother’s age at child’s birth

  5. Maternal highest level of education

  6. Paternal highest level of education

  7. Wealth index

  8. Neonatal mortality

  9. Post-neonatal mortality

  10. Infant mortality

  11. Child mortality

  12. Under 5 mortality

The study used pooled analysis of 2004, 2007
and 2011 Bangladesh DHS with an average response rate of 98%.
The researchers used multilevel modelling, which took into account the effect of clustering to better estimate the level of association of the study factors with the outcome
Mother’s working status OR for working/non-working status for each type of mortality: 1.35, 95% CI 1.01 to 1.80, p=0.041 for neonatal mortality; 1.90, 95% CI 1.32 to 2.74, p=0.001 for postnatal mortality; 1.60, 95% CI 1.26 to 2.01, p<0.001 for infant mortality; 1.67, 95% CI 1.34 to 2.08, p<0.001 for under 5 mortality.
Maternal highest education: AOR for No education/primary/secondary for each type of mortality: 1.00/0.79, 95% CI 0.59 to 1.04, p=0.096/0.51, 95% CI 0.32 to 0.83, p=0.007 for neonatal mortality; 1.00/0.81, 95% CI 0.54 to 1.46, p=0.280/0.28, 95% CI 0.10 to 0.78, p=0.015 for post-neonatal mortality; 1.00/0.80, 95% CI 0.63 to 1.02, p=0.069/0.45, 95% CI 0.29 to 0.70, p<0.001 for infant mortality; 1.00/0.83, 95% CI 0.66 to 1.04, p=0.104/0.41, 95% CI 0.26 to 0.63, p<0.001 for under 5 mortallity.
Contraceptive use: OR for no/yes use of contraception for each type of mortality: 1.00/0.30, 95% CI 0.23 to 0.39, p<0.001 for neonatal mortality; 1.00/0.49, 95% CI 0.35 to 0.70, p<0.001 for post-neonatal mortality; 1.00/0.35, 95% CI 0.28 to 0.43, p<0.001 for infant mortality; 1.00/0.22, 95% CI 0.11 to 0.42, p<0.001 for child mortality, 1.00/0.33, 95% CI 0.27 to 0.40, p<0.001 for under 5 mortality
Maternal working status increases the risk of neonatal to under 5 mortality.
Secondary education of mother significantly reduces the risk of neonatal to under 5 mortality.
Contraceptive use significantly decreases the risk of neonatal to under 5 mortality
Emenike et al (2008)
Kenya30
Describe the association of IPV on the reproductive health of womenAll women 15–49 years of age residents or visitors at the sampled household at the time of the survey were eligible for participation. Only women ever (currently or formerly) married/having a partner and who responded to the domestic violence module (n=4312) were included from Kenya DHS 2003
  1. Family planning method

  2. Terminated pregnancy

  3. Infant mortality

  4. Number of births ever

  5. Intimate partner violence

Cross-tabulation was used to study the association between the dependent and independent variables, and significant levels were tested using χ2 test. Because age may be associated with both reproductive health and IPV, age-adjusted associations between IPV and reproductive health indicators were calculated using logistic regression analysesPhysical IPV/family planning OR 1.236, 95% CI 1.086 to 1.406, p<0.01; Physical IPV/infant mortality OR 1.652,
95% CI 1.432 to 1.906, p<0.01


Emotional IPV/family planning OR 1.419, 95% CI 1.221 to 1.560, p<0.01; Emotional IPV/infant mortality OR 1.483,
95% CI 1.226 to 1.736, p<0.01;
Sexual IPV/family planning OR 1.684, 95% CI 1.390 to 2.040, p<0.01; Sexual IPV/infant mortality OR 1.250,
95% CI 1.029 to 1.520, p<0.01
Use of family planning methods and infant mortality was significantly associated with physical/emotional or sexual violence.
The results also suggest a strong association between the use of family planning methods and increased vulnerability to IPV
Okenwa et al (2010)
Nigeria31
Examine the association of IPV on reproductive health outcomes33 385 women aged 15–49 years, including permanent residents of the households or visitors present in the households on the night before the survey, were included from Nigerian DHS 2008
  1. Abortion

  2. Contraceptive use

  3. Pregnancy wish

  4. Exposure to physical IPV

  5. Exposure to mental IPV

  6. Exposure to sexual IPV

  7. Education

  8. Religion

  9. Place of residence

  10. Region

The χ2 test was used to test for associations between IPV and the independent variables in the univariate analyses. The independent association between IPV exposure and the dependent variables (after control for potential confounding) was determined using logistic regression.
Direction and magnitude of the associations were expressed as adjusted OR
(AOR)
Physical IPV/contraceptive use No vs Yes AOR 0.792, 95% CI 0.687 to 0.912, p=0.001; Physical IPV/infant mortality OR 0.846, 95% CI 0.771 to 0.928; p=0.000;
Emotional IPV/contraceptive use No vs Yes AOR 0.852, 95% CI 0.743 to 0.977, p=0.022; Emotional IPV/infant mortality OR 0.755, 95% CI 0.699 to 0.819, p=0.000;
Sexual IPV/contraceptive use No vs Yes OR 0.952, 95% CI 0.708 to 1.282, p=0.748; Sexual IPV/infant mortality OR 0.798, 95% CI 0.670 to 0.951, p=0.011
Contraceptive use was significantly associated with physical and emotional violence (p value for sexual violence was insignificant).
Physical, emotional and sexual IPV was significantly associated with infant mortality
  • AOR, Adjusted Odds Ratio; CI, Confidence Interval; DHS, Demographic and Health Survey; IPV, intimate partner violence; OR, Odds ratio.