Table 1.

Summary of included trials

Study	Location	Primary focus	Intervention	Control	Sample size	Inclusion criteria	Exclusion criteria	Key results	Limitations
Cramp and Brawley (2006, 2009)	Canada	Physical activity (PA)	Group-mediated cognitive behavioural intervention (six sessions immediately following standard postnatal exercise class) at a community-based fitness facility. Focus on developing self-regulatory skills for self-management of physical activity and overcoming postnatal barriers to physical activity	Standard postnatal exercise training programme	57	Women with a (1) primarily sedentary lifestyle, (2) up to a year after giving birth	(1) Long-term medical conditions, (2) currently pregnant, (3) non-English speaking	(1) Group participants reported significantly higher change in frequency, minutes, and volume of PA, (2) group participants’ expectations of the likelihood of achieving their proximal outcomes remained stable, while expectations in the control arm decreased, (3) mean barrier efficacy increased and self-regulatory efficacy remained stable for group participants, while both declined in the control group	(1) Small sample size, (2) participants were self-selected so perhaps unrepresentative of the general postnatal population, (3) participation in PA was self-reported, (4) no follow-up was conducted beyond the length of the intervention itself
Escobar et al. (2001)	US	Health service use and quality of care	Home visits by a nurse (within 48 h after hospital discharge) NB: for the purposes of this review we are interested in the control arm, which includes a group-based intervention	Hospital-based group postnatal education within the first 72 h. The session for up to eight mother–infant pairs was led by a nurse and included physical assessment of the baby, advice on breastfeeding and basic infant care and preventative education	1014	(1) Low medical and social risk (as defined by pre-determined criteria), (2) hospital length of stay was expected to be ≤ 48 h	N/A	(1) No significant differences occurred with respect to: maternal urgent visits, neonatal urgent visits, maternal or neonatal rehospitalisation, breastfeeding discontinuation, or the occurrence of maternal depressive symptoms, (2) mothers in the home visit group were more likely than those mothers in the hospital-based care group to rate multiple aspects of their care as excellent or very good	(1) Studied a socioeconomically low-risk population, (2) only one group session offered, (3) limited power to establish differences between study arms for individual outcomes, (4) poor response rates to questionnaire about maternal satisfaction
Hagan et al. (2004)	Australia	Postnatal depression	A 6-session cognitive behavioural group therapy intervention programme delivered by a research midwife between weeks 2 and 6 after birth	Standard carea	199	English-speaking mothers of preterm babies (< 33 weeks) admitted to a neonatal unit	Among others (1) age < 17 years, (2) diagnosis of psychotic disorders, and/or depression, and/or significant substance abuse	No differences between study arms in new diagnoses of depression (using DSM-IV criteria) in the 12 months after delivery	(1) Only half of eligible women participated in the trial, (2) intervention sessions were held when most mothers were still caring for their infant in the neonatal unit
Reid et al. (2002)	Scotland	Physical and mental health, and health service use	Multi-factorial design with two interventions: (1) “Pack”: a postnatal self-help manual, and (2) “Group”: midwife-facilitated weekly support groups, held at community centres. The session agenda was decided with attendees	Not detailed	1004	Among others (1) primiparous women, (2) attending one of two Scottish maternity hospitals	(1) Death of the infant, or (2) admission to SCBUb for > 2 weeks	No significant differences between study arms in scores for three standardised screening tools for postnatal depression, general health and social support. Also no significant differences in health service use	(1) Poor intervention uptake, (2) group attendees were more likely to be middle class and own their home
Rouhe et al. (2015)	Finland	Life satisfaction, wellbeing and healthcare costs for women with fear of childbirth	Group psycho-education led by a psychologist: six sessions during pregnancy and one after delivery	Women received a letter advising them to discuss their fear of childbirth with their usual care provider	371	(1) Nulliparous women, (2) severe fear of childbirth, (3) during first trimester of pregnancy	Not detailed	(1) The groups did not differ in total direct costs, nor in life satisfaction or general wellbeing, (2) SVDc with no complications was registered more often in the intervention group, (3) numbers of non-complicated and complicated caesarean section were higher in the control group	(1) The return rate of self-report questionnaires was low, (2) only one of the sessions took place in the postnatal period, (3) data were not detailed enough to enable cost analysis of antenatal admissions, screening or induction of labour
Ryding et al. (2004)	Sweden	Women after emergency caesarean	Group counselling: two sessions at two months postpartum, with four to five participants. Facilitated by a psychologist and a midwife	Standard care	162	Women giving birth to a live infant by emergency caesarean section at Helsingborg Hospital in Sweden	Non-Swedish speaking women	No difference between groups was found at 6 months postpartum for (1) frightening memories of childbirth, (2) symptoms of posttraumatic stress, (3) postnatal depression	(1) Small sample size, (2) only two intervention sessions offered
Stamp et al. (1995)	Adelaide, South Australia	Postnatal depression	Two antenatal groups (at 32 and 36 weeks gestation) and a postnatal group (6 weeks postpartum) by led a midwife educator. Partners were welcome to attendd	Not detailed	144	Women with (1) a singleton pregnancy < 24 weeks’ gestation, (2) vulnerable to postnatal depression, (3) living within the metropolitan area	(1) Non-English speaking women, (2) privately insured women not attending the clinic	No significant difference between the groups’ postnatal depression scores at 6 and 12 weeks and 6 months postpartum	(1) Small sample size, with limited power to detect small changes in depression score, (2) only one of the sessions took place in the postnatal period, (3) poor attendance at groups, particularly in the postnatal period
Tandon et al. (2011, 2014) and Mendelson et al. (2013)	US	Postnatal depression and mood regulation	Standard home visiting services plus the Mothers and Babies (MB) Course: a cognitive behavioural intervention, consisting of six group sessions led by a clinical social worker or clinical psychologist	Standard home visiting services plus information on perinatal depression	78	Women who were (1) pregnant or who had a child less than 6 months of age, (2) enrolled in one of four home visiting programs, (3) at risk for perinatal depression	Women with a current depressive episode	(1) Depressive symptoms declined at a greater rate for the intervention group at 1 week, 3 months and 6 months. However no significant difference was reported between depressive episodes at 6 months, (2) the intervention group experienced greater growth in mood regulation from baseline to 6-month follow-up	(1) Small sample size, with limited power to detect small effects, (2) 28% of women were pregnant at baseline (i.e., at least some of intervention sessions took place antenatally), (3) structured clinical interview to diagnose depressive episode was only conducted at the final 6-month follow-up
Wiggins et al. (2004, 2005)	UK	Several maternal and child health outcomes	(1) Support health visitor (SHV) intervention: seven home visits and additional telephone contacts, (2) community group support (CGS) intervention: community groups for mothers with children < 5 years offering: drop-in sessions, home visiting, and/or telephone support	Standard health visitor services	731	Women living in the Inner London Boroughs of Camden and Islington	Women whose babies (1) had died, (2) were seriously ill, (3) had been placed in foster care	(1) At 12 and 18 months, no significant differences were found between the interventions with respect to: child health, child injury requiring medical attention, infant feeding, maternal smoking levels, and prevalence of maternal depression, (2) SHV women had different patterns of health service use (with fewer taking their children to the GP), and (3) had less anxious experience of motherhood than control women	(1) Poor uptake of services in CGS arm, (2) power to detect effects was limited by presence of two study arms, (3) women who declined to participate were more likely to be from an ethnic minority, (4) uptake of both interventions was lower for women whose first language was not English, (5) the standardised tools used had not been validated in the cultural groups involved in the study

^aThis study included regular biweekly education group sessions

^bSpecial care baby unit

^cSpontaneous vaginal delivery

^dThese groups were offered in addition to the hospital’s antenatal classes