Table 2 Overview of Included Studies

Ayiasi, 2016, Uganda

Pregnant women (Masindi & Kiryandongo districts)

Cluster RCT

Treatment (n=627): In addition to routine educational messages in ANC clinic, received village health teams (VHTs) making home visits to provide educational messages for maternal/newborn care & each VHT had mobile phone handset capable of making unlimited phone consultation with health workers; VHTs made 2 ANC visits (4 weeks apart) & 1 PNC home visit 3 days after delivery. Control (n=758): Received group education routinely offered in health centers, but did not receive VHT home visits or mobile phones. Follow-up: Data were collected 4-6 months after delivery.

Primary: Health facility delivery. Secondary: ANC attendance, birth preparation, cord care, thermal care, initiation of exclusive breastfeeding, & care-seeking for newborn illness. ANC attendance was measured through self-report where attending three or more ANC visits was categorized as adequate & the rest were grouped as inadequate.

85% of the intervention group made 3+ antenatal visits, compared to 71% of the control, adjusted odds ratio 1.82 (95% CI 0.65-5.09, p=0.26)

Home visit intervention had no effect on recommended ANC attendance.

Cherniak, 2017, Uganda

Pregnant women (Kabale district)

Cluster RCT

Treatment (n=100): Word of mouth ANC advertisement carried out by local community leaders who announced free ANC at community gatherings, plus advertisement for availability of portable ultrasound (pOBU), further divided into word of mouth of pOBU & ANC (n=16), radio advertisement of only ANC & word of mouth of ANC & pOBU (n=7), or word of mouth plus radio of both ANC & pOBU (n=75). Control (n=59): Word of mouth advertisment of ANC only with no mention of pOBU. Follow-up: Data were collected upon arrival to clinic and leaving clinic.

Primary: ANC attendance/uptake. Secondary: Attendance by women who previously used a traditional healer/birth attendant (TBA), had not yet attended ANC, attendance between three interventions, & # of women stating they came for pOBU. ANC uptake rate calculated using # of women attending ANC as the numerator & number of women attending first ANC in 2013-2014 through government-run clinics as the denominator.

Rate of ANC attendance was 65.1% per 1000 pregnant women where pOBU advertised by radio & word of mouth vs. 11.1% in control communities, rate ratio 5.86 (95% CI 2.6-13.0, p<0.0001)

Advertising pOBU by radio messaging significantly increased ANC uptake as compared to word of mouth advertisement of ANC only.

Kirkwood, 2013, Ghana

Pregnant women (Brong Ahafo region)

Cluster RCT

Treatment (n=9174): Integrated intervention training community based surveillance volunteers (CBSVs) to identify pregnant women in their community, undertake 2 home visits during pregnancy & 3 visits after birth on days 1, 3, & 7 in addition to standard care provided. CBSVs also were responsible to weigh the newborn & check them for danger signs after birth. Control (n=9435): Standard care available, including antenatal clinics, access to free facility delivery, post-partum checkups, infant welfare clinics, & routine CBSV activities for outreach. Follow-up: Data were collected via surveillance home visits that occurred every 4-8 weeks throughout the study period.

Primary: All-cause neo-natal mortality rate (NMR) in the 1st 28 days of life & % of mothers practicing Newhints recommended behaviors (including # of ANC visits). Secondary: Age-specific & cause-specific NMRs. Attendance to 4 or more ANC visits was measured via self-report following birth.

76% treatment & 73.7% control attended 4 or more ANC visits (2.3% incr.), relative risk 1.02 (95% CI 0.96-1.09, p=0.52)

Home visit intervention had no effect on recommended ANC attendance.

Magoma, 2013, Tanzania

Pregnant women (Ngorongoro district, Arusha region)

Cluster RCT

Treatment (n=404): Introduction & promotion of birth plans by care providers during ANC to prepare women & families for birth. Discussions on place of delivery, importance of skilled delivery care, transport arrangements, funding, possible blood donors, birth companions, & home support. Control (n=501): Standard care without birth plan. Follow-up: Data were collected within 1 month of delivery.

Primary: Delivery in health unit. Secondary: PNC attendance, satisfaction of women & providers with care received & provided. PNC attendance within 1 month of delivery was determined from self-reports & cross-checked using health records.

PNC utilization within 1 month: 62.1% in treatment & 32.1% in control, adjusted absolute difference 31.3% (95% CI 15.4-47.2, p=0.0009); days to initial PNC (mean +- SD) treat 6.6 +-1.7 vs. control 20.9+-4.4, p=0.0001

Introduction & promotion of birth plans during ANC care increased PNC utilization in the first month after delivery.

Waiswa, 2015, Uganda

Pregnant women (Iganga & Mayuge districts)

Cluster RCT

Treatment (n=894): 5 home visits by community health workers, 2 during pregnancy & 3 in the 1st week after birth (day 1, 3, & 7) to offer preventative & promotive care/counseling with extra visits for sick & small newborns to assess & refer plus improved facilities Control (n=893): Standard care overseen by district health team in addition to improved facilities. Follow-up: Endline data were collected amongst women who had a live birth within 12 months of the baseline survey.

Primary: Coverage of ANC, birth preparedness, skilled attendance at delivery, PNC, breastfeeding, thermal care, & hygiene. Secondary: None reported. Data on attendance to one or more ANC visits and to four or more ANC visits were collected via self-report.

99.2% intervention & 98.9% control attended at least one ANC visit, p=0.44); 47% intervention & 43.6% control attended 4 or more ANC visits, p=0.165

Home visit intervention had no effect on recommended ANC attendance.