Table 2

Overview of studies in systematic review (n=14)

Author (date)	Country	Programme details	Evaluation details	Findings
Universal unconditional programmes (n=3)
Forget25	Canada	MINCOME: a mid-1970s experimental guaranteed annual income programme with transfer amounts based on a sliding scale (depending on income from other sources). Recipients: All residents of Dauphin, Manitoba.	Design and sample size: Families were randomly selected to participate from Dauphin and were matched to controls from similarly small rural communities. Data source: Manitoba Population Health Research Data Repository.	No significant differences in perinatal death and birth weight (estimates not provided).
Li and Mora29	Nepal	Old Age Allowance Programme (OAAP): a universal flat rate pension Recipients: Residents aged 75 years or older.	Design and sample size: Treated group: OAAP eligible person in same household as infant (n=211). Controls: those not living in household with OAAP eligible person (n=13 345). Data source: Nepal Demographic and Health Survey from 1996 to 2001.	Households with an OAAP eligible elderly person, OAAP resulted in 7%–8% higher probability of survival 12 months after birth (from a baseline probability of 89%).
Chung et al 38	USA	Alaska Permanent Fund Dividend (APFD): an annual dividend derived from a state fund, which accumulates through the sale of oil drilling rights. On average, families with newborns received US$1961. Recipients: With the exception of having being a resident of Alaska for at least 6 months, an equal dividend is given unconditionally to every resident of the state.	Design and sample size: Difference-in-differences modelling to determine the impact of APFD on birth outcomes in Alaska compared with control states with no such fund (n=7.7 million) Data source: Vital Statistics (the US Natality Detail File from 1978 to 1984).	US$1000 of APFD was associated with a 17.7 g increase in birth weight, and a decrease in between 0.4%and 0.7% in low birth weight. Apgar scores also showed improvement. Effects were largest for those with a high school diploma or less.
Targeted unconditional programmes (n=3)
Brownell et al 24	Canada	Healthy Baby Prental Benefit (HBPB): an income support of up to US$81.41 per month. Recipients: Women in Manitoba, Canada during second and third trimesters, with a documented income below US$32 000 per year.	Design and sample size: All infants born to Manitoba women from 2003 through 2010 on welfare during pregnancy (total n=15 033: those receiving HBPB (n=10 738) and those not receiving HBPB (n=3853). Data source: Administrative linked data.	Receiving HPBP associated with: 29% reduction in risk of low birth weight (95% CI 0.63 to 0.81), 24% reduction in risk of preterm birth (95% CI 0.69 to 0.84), 10% reduction in risk of small for gestation age (95% CI 0.81 to 099) and 13% increase in large for gestational age (95% CI 1.05 to 1.23). Population Attributable Fractions suggested HBPD associated with 21% reduction in low birth weight and 17.5% reduction in preterm birth in the population.
Currie and Cole22	USA	Aid to Families with Dependent Children (AFDC): a cash transfer to single mothers. Recipients: 11 million people in 1988.	Design and sample size: n=approximately 5000 children born between 1979 and 1988 with complete AFDC and birth weight data. Data source: National Longitudinal Survey of Youth (NLSY)	No significant association between AFDC and birth weight. Marginal association with increased birth weight among poor whites (beta=32.00, SE: 16.11).
Kehrer and Wolin23	USA	Gary Income Maintenance Experiment: a federally sponsored test of negative income tax between 1971 and 1974. Recipients: Urban, Black, low-income population with high concentration of female-headed households in Gary, Indiana.	Design and sample size: Treatment families were those eligible to receive various levels of income maintenance (n=256). Control families were from Gary, Indiana, but not eligible for income maintenance (n=148). Data source: Birth records of Gary Income Experiment participants and monthly income reports.	Children in the treatment group experienced 117.6–530.4 g higher birth weight than children in the control group.
Targeted conditional programmes (n=8)
Barber and Gertler27	Mexico	Oportunidades (formerly Progresa): a conditional cash transfer programme based on age, including prenatal care, healthcare use and school attendance. Recipients: Approximately 5 million low-income families in urban and rural settings.	Design and sample size: Two-staged random, probability-based sample of communities and households within communities. Total n=840 women (n=666 beneficiary births, n=174 non-beneficiary births). Data source: Fertility surveys and household interviews	Beneficiary status associated with 127.3 g higher birth weight (95% CI 21.3 to 233.1) and 4.6% decrease in low birth weight.
Barham26	Mexico	Oportunidades (formerly Progresa): a conditional cash transfer programme based on age, including pre-natal care, healthcare use and school attendance. Recipients: Approximately 5 million low-income families in urban and rural settings.	Design and sample size: Randomisation of 506 programme localities in seven states into treatment and control groups. Data source: Mexican vital statistics data from 1992 to 2001.	Oportunidades was associated with a reduction in 3 deaths per 1000 live births, or 17% decline (P=0.01). No significant association effect of Progresa on neonatal mortality rate.
Shei28	Brazil	Bolsa Família: monthly cash transfers attached to health (routine growth monitoring, vaccinations, regular prenatal care) and educational (enrolment in school, minimum attendance requirements) conditionalities. Recipients: Poor households with per capita incomes below a program-specific poverty line.	Design and sample size: Pooled, time series, cross-sectional design; sample sizes not provided. Data source: Brazilian Unified Health System Database, Ministry of Social Development and Brazilian Institute of Geography and Statistics, at the municipal level, from 1998 to 2008.	While IMR was already falling prior to 2003, rate of decline increased after programme implementation. Average treatment effect was 9.3% decline in IMR and 24.3% decline in postneonatal IMR (P<0.01), but no significant change in neonatal IMR.
Bruckner et al 21	USA	Earned Income Tax Credit (EITC): tax refund disbursed to low-income families contingent on employment, with larger benefits to families with children. Credits increase with increasing income until a maximum threshold credit is reached (US$4716). Median refund was US$1130 during the study period. Recipients: 95% of recipients of EITC were single or married couples with a qualifying child under the age of 19 years.	Design and sample size: Used variables from the California Birth File to specify (1) gravid women who might have been eligible for an EITC benefit and (2) their children born during the study period (n=70 895). Data source: California Birth File from 1989 to 1997 (period of high EITC expansion and data consistency)	Odds of very low birth weight increased 2 months immediately following EITC disbursement (OR 1.31, 95% CI 1.09 to 1.58). Effect of EITC lagged at 3 or 4 months was also non-significant.
Cho18	USA	Personal Responsibility Work Opportunity and Reconciliation Act: ended the federal guarantee of income support, imposed lifetime limits on public assistance, created additional work requirements and allowed states to tie assistance to specified maternal behaviours. Recipients: Legislation largely affected poor families with children, especially immigrant families.	Design and sample size: Total n=881 854 (n=562 937 foreign-born Mexican women and n=318 917 comparison group native-born Mexican women). Data source: National Centre for Health Statistics infant birth and death records on low-income Mexican women from 1995 to 1996 (before welfare reforms) and 1999–2002 (after welfare reforms).	Welfare reform was associated with an increase in infant mortality rates for foreign-born Mexican women by 3.1 deaths per 1000 live births (SE: 0.001). In a more socioeconomically restricted sample, welfare reform was associated with an increase of 5.9 deaths per 1000 live births (SE: 0.002).
Hamad and Rehkopf20	USA	EITC: tax refund disbursed to low-income families contingent on employment, with larger benefits to families with children. Credits increase with increasing income until a maximum threshold credit is reached (US$4716). Recipients: 95% of recipients of EITC were single or married couples with a qualifying child under the age of 19 years.	Design and sample size: Multivariable regression and intstrumental variable analysis of study sample women (n=2985) and their children born during 1986–2000 years during which ETIC payment size varied the most (n=4683). Data source: NLSY.	Each additional US$1000 of EITC payment was not significantly associated with an increase in birth weight of 65.1 g (95% CI −0.46, 130.6). Results of instrumental variable analysis suggest no significant effect of EITC in the overall sample, but an increase in birth weight for women receiving the payment in the second trimester.
Strully et al 19	USA	EITC: tax refund disbursed to low-income families contingent on employment, with larger benefits to families with children. Credits increase with increasing income until a maximum threshold credit is reached (US$4716). Recipients: 95% of recipients of EITC were single or married couples with a qualifying child under the age of 19 years.	Design and sample size: Differences-in-differences modelling with sample limited to unmarried mothers with a high school degree or less as a way to capture those most likely to receive EITC. Data source: Vital Statistics (the US Natality Detail File from 1980 to 2002).	EITC was associated with an average increase in birth weight of 15.70 g (SE: 1.211) among unmarried women with high school education or less.
Hoynes et al 39 (2015)	USA	EITC: tax refund disbursed to low-income families contingent on employment, with larger benefits to families with children. Credits increase with increasing income until a maximum threshold credit is reached (US$4716). Recipients: 95% of recipients of EITC were single or married couples with a qualifying child under the age of 19 years.	Design and sample size: Differences-in-differences modelling of three EITC time ranges for single mothers aged 18 years or older with less than a high school education singleton births (n=35 467–47 687) Data source: Vital statistics (the US Natality Detail File from 1984 to 1998 and Current Population Survey.	US$1000 of EITC was associated with a 2%–3% decline in low birth weight and an increase of 6.4 g in mean birth weight. Preterm birth, small for gestational age and Apgar score also improved. Effects were largest for black infants and smallest for white infants.