Table 1 Description of included studies

Aquino 2009 [23] Brazil

Interventions to improve provision of maternal and child health care services: Family Health Program (Strengthening community health service provision)

MDG 4: Infant and under-five mortality

Ecological study

Socioeconomic status

Fixed-Effects Models stratified by human development index of a district; Reductions in infant mortality rates measured by Risk Ratios after Family Health Program being more in lower Human development index group

Moderate

Baqui 2008 [24] India

Interventions to improve provision of maternal and child health care services: NGO facilitation of a government community-based MCH program

MDG 5: Skilled birth attendance and antenatal care visit

Controlled before and after study

Asset/wealth

Concentration indices (CI) calculated for intervention and comparison districts at baseline and endline for each outcome; The change of CI from baseline to endline in intervention districts being significant, but not in comparison districts

Strong

Benn 2010 [6] Guinea-Bissau

Nutrition supplements: Vitamin A

MDG 4: Infant and under-five mortality

Controlled trial

Gender

The differences in mortality between Vitamin A group and Placebo group, and stratification by sex; mortality rate ratios of vitamin A supplementation showing the supplementation tended to be beneficial in boys but not in girls

Strong

Bishai 2002 [25] Bangladesh

Expanding immunization coverage campaign: Intensive outreach visits by community health workers

MDG 4: Measles vaccination

Cohort study

Education, size of living room

Interactions between socioeconomic status indictors and a dummy variable for residence in intervention areas was added in probit regressions; The effects of parents’ education and living size being both lessened in intervention areas

Strong

Bishai 2003 [26] Bangladesh

Expanding immunization coverage campaign: Measles vaccine

MDG 4: Infant and under-five mortality

Cohort study

Socioeconomic status

Socio-economic differentials in mortality between the lowest and highest socioeconomic status (SES) quintiles in a cohort of vaccinated children and a control cohort; The mortality ratio of lowest SES to highest in the control population being significantly higher than it in the vaccinated population

Strong

Bishai 2005 [27] Nepal

Nutrition supplements: Vitamin A

MDG 4: Infant and under-five mortality

Controlled trial

Gender, caste and asset

The differences in mortality between boys and girls in Vitamin A group and placebo group; The difference in Vitamin A group being less than the difference in placebo group

Strong

Dyer 1996 [28] South Africa

Expanding immunization coverage campaign: National immunization campaign

MDG 4: Measles vaccination

Repeated cross-sectional study

Ethnicity

The differences vaccination coverage between pre- and post- campaign, and stratified for race; The difference in Asians being more than the differences in urban and rural blacks

Weak

Hatt 2007 [29] Indonesia

Interventions to improve provision of maternal and child health care services: A midwife in every village

MDG 5: Skilled birth attendance

Repeated cross-sectional study

Asset/wealth

Time-trend interactions with wealth quintile and urban/rural residence was added into logistic regression; The Odd Ratios showing that professional attendance in low socioeconomic status group increased more than those with higher socioeconomic status

Weak

Hotchkis 2011 [30] Nigeria, Uganda, Bangladesh and Indonesia

Interventions to improve provision of maternal and child health care services: Expansion of the private sector supply of modern contraceptives

MDG 5: Modern contraceptive prevalence Rate

Repeated cross-sectional study

Asset/wealth

Concentration indices differences between different survey rounds; concentration indices overall showing decrease trend

Moderate

Koenig 2001 [31] Bangladesh

Expanding immunization coverage campaign: Measles vaccine

MDG 4: Infant and under-five mortality

Cohort study

Gender, household dwelling size and education levels

Interactions between socioeconomic status indictors and vaccination status was added in proportional hazard regression; The decline in mortality risks with measles vaccination for those from smaller household dwelling size significantly exceeding that for those with bigger dwelling size, but no significant results for other two dimensions (gender and education level)

Moderate

Kozhimannil 2009 [32] Phillipines

Demand side interventions: Health insurance program

MDG 5: Antenatal care attendance

Repeated cross-sectional study

Asset, occupation and place of residence (rural/urban)

Stratified logistic analyses by urban/rural, employment status, and wealth; In women in rural areas, those in the lowest wealth tertile and those who are employed, the odds ratios being bigger and significant

Moderate

Masanja 2005 [33] Tanzania

Interventions to improve provision of maternal and child health care services: Integrated management of Childhood Illness

MDG 4: Measles vaccination

Controlled before and after study

Socioeconomic status

Concentration indices differences from 1999 to 2002 in both intervention and control groups; The difference in concentration indices in intervention group reducing more

Moderate

Nasreen 2003 [34] Bangladesh

Interventions to improve provision of maternal and child health care services: BRAC intervention (Providing preventive health, nutrition education and other primary care)

MDG 4: Infant and under-five mortality

Case control study

Education, occupation, economic condition, age

Stratified analysis for association between BRAC membership and neonatal death; For those mothers aged less and those fathers without wage, the relative risks of death being greater in intervention groups, and no significant influence from economic condition and parents’ literacy

Strong

Nielsen 2005 [35] Guinea-Bissau

Nutrition supplements: Vitamin A

MDG 4: Infant and under-five mortality

Controlled before and after study

Education levels, ethnicity, place of residence

Changes in mortality related to socio-economic factors for Vitamin A supplemented children compared with pre-wartime (no Vitamin A supplementation); the mortality risks being significantly reduced after war (Vitamin A supplemented) in the area pre-war disadvantage of families living and for those with low mother education level

Moderate

Pebley 1991 [36] Indiana

Mixed interventions: Combined health-care and nutrition service provision

MDG 4: Infant and under-five mortality

Controlled trial

Gender

The changes in male/female mortality ratios from before to during the intervention in different interventions and control groups; the male/female mortality ratios being increased more in intervention groups than change in control group

Weak

Sasaki 2011 [7] Zambia

Expanding immunization coverage campaign: Outreach immunization services

MDG 4: Measles vaccination

Repeated cross-sectional study

Distance to health facilities and income

Two logistic regression analysis before and after the introduction of outreach immunization services; Before the introduction of outreach services, longer distances to the service points being associated with lower measles immunization coverage, but not after the outreach services; no significant outcome for income

Moderate

Zeng 2011 [37] China

Nutrition supplements: Multiple micronutrients

MDG 4: Infant and under-five mortality

Controlled trial

Asset/wealth

Analysis for association between interventions and mortality outcome stratified by household wealth index; in the poorest group Iron/folic acid supplement significantly being related lower early neonate mortality, the association being not significant in wealthier group

Strong

Kamiya 2013 [38] Bangladesh

Mixed interventions: Safe Motherhood Promotion Project including demand side intervention (community mobilization through participatory approaches) and intervention to improve service provision

MDG 5: Skilled birth attendance and antenatal care visit

Controlled before and after study

Income

Interaction term, Income quartile × Project × Time was added in Difference-in-differences logistic regressions; Relative to the comparison site, significant increase in antenatal care for women belonging to lower income quartiles compared to those in the highest quartile in the project site, no significant outcome for skilled birth attendance

Strong

Quayyum 2013 [39] Bangladesh

Interventions to improve provision of maternal and child health care services: BRAC program’s the intervention on improving Maternal, Neonatal and Child Survival (IMN CS)

MDG 5: Skilled birth attendance and antenatal care visit

Controlled before and after study

Asset/wealth

The change in concentration index over time in both intervention and comparison areas; The concentration index being reduced more in the intervention areas

Strong

Amudhan 2013 [40] India

Mixed interventions: demand side interventions (conditional cash transfer scheme) and interventions to improve services provision (the strengthening of the primary health centre network)

MDG 5: Skilled birth attendance

Controlled before and after study

Socio-economic status: caste and education

Subgroup analysis by socioeconomic status on institutional delivery rate; institutional delivery among disadvantaged mothers increasing more compared with the increase among other groups

Moderate

Houweling 2013 [41] Indian

Demand side interventions: Participatory women’s group intervention

MDG 4: Infant and under-five mortality

Controlled before and after study

Socio-economic status: caste, land ownership, literacy, and asset ownership

Separate random effects logistic regression, for the most and less socio-economically marginalized groups; Odd Ratios showing that mortality reducing more in most marginalized populations than the reduction in less marginalized populations

Strong

Mosquera 2012 [42] Colombia

Interventions to improve provision of maternal and child health care services: Primary Health Care (PHC) strategy

MDG 4: Infant and under-5 mortality rate

Ecological study

Socio-economic status

Change in concentration indices for four child health outcomes from before to after implementation of intervention; The variation in the concentration index observed between the two periods being positive

Weak