Table 1 Characteristics of studies on newborn care providers in low-and middle-income countries

Ansah et al. 2014 [20]

Ghana

Community based-surveillance volunteer (CBSV)

Training program for home visits by CBSVs

9 days

3 days on behaviour change communication, counseling skills, promotion of essential newborn care practices; 4 days on assessment and referrals; 2 days of refresher course and clinical practice sessions in major health facilities.

Yes

Not assessed

Not assessed

Assessed

Not assessed

Not assessed

CBSVs’ newborn assessments strongly agreed with district-based project supervisors (DiPS) with coefficients of agreement between 0.75–1.0. The sensitivities of CBSVs’ diagnosis for signs checked by observation were relatively low (57–59%) with just >40% detected by the DiPS missed by the CBSV. However, specificities were close to 100% for all danger signs.

Referral decisions made by the CBSVs also achieved excellent agreement with the DiPS; Kappa = 0.87 (0.82, 0.92), with 80% sensitivity and 100% specificity.

Bang et al. 2005a [21]

India

Village health worker (VHW)

Health education, including newborn danger signs

6 months

Taking histories of pregnant women, observing labor, examining neonates, recording findings, case management of pneumonia in neonates, supporting mothers with home neonatal care and identifying neonatal morbidities.

Yes

Not assessed

Assessed

Not assessed

Assessed

Not assessed

Regarding caregiver knowledge of danger signals in a baby when the VHW should be called, 77.3% of mothers correctly responded.

Regarding caregiver identification of newborns with serious illness, 87.2% of mother correctly called a VHW if a baby was sick.

Bang et al. 2005b [22]

India

VHW

VHW home visits

2 months

Taking a history, examining a mother and newborn, and recording data. Trained to give intramuscular vitamin K, diagnose and treat sepsis.

Yes

Not assessed

Not assessed

Assessed

Not assessed

Assessed

VHWs correctly diagnosed 492 (89.1%) of neonatal sepsis cases compared to a computer algorithm.

Of the 492 patients diagnosed by the VHW to have presumed sepsis, parents agreed to and were able to hospitalize only 13 infants (2.6%), but agreed to homebased treatment for almost all infants (91.1%). Of note, in 31 cases (6.3%), parents refused both home and hospital care.

Bang et al. 2005c [23]

India

VHW

VHW home visits

2 months

Taking a history, examining a mother and newborn, and recording data. Trained to give intramuscular vitamin K, diagnose and treat sepsis.

Yes

Not assessed

Not assessed

Assessed

Assessed

Not assessed

Among LBW or preterm infants or both, VHWs correctly diagnosed sepsis 94% of the time and correctly treated sepsis 95% of the time.

Baqui et al. 2009a [24]

Bangladesh

Community health worker (CHW)

Performance of CHWs in assessing neonates using an Integrated Management of Childhood Illness (IMCI)-type algorithm, within the home care arm of the trial

6 weeks

Skills development for behaviour change communication, clinical assessment of neonates, treatment of newborns with injectable antibiotics and record-keeping; hands-on clinical training under supervision in a tertiary care hospital and in households.

Yes

Not assessed

Not assessed

Assessed

Not assessed

Not assessed

Compared to physician assessment, CHWs are able to correctly classify very severe disease in newborns with a sensitivity of 91%, specificity of 95%, and kappa value of 0.85 (p < 0.001). CHW recognition of newborn signs and symptoms showed a sensitivity of more than 60% and a specificity of 97–100%.

Baqui et al. 2009b [25]

Bangladesh

CHW

Analysis of data in home care arm (see Baqui et al. 2009a)

6 weeks

As above.

Yes

Not assessed

Not assessed

Assessed

Not assessed

Assessed

CHW home-based treatment of very severe disease in neonates was acceptable and resulted in a hazard ratio for death of 0.22 (95%CI 0.07–0.71) compared to those who received no treatment or were treated by untrained providers.

Bari et al. 2006 [26]

Bangladesh

CHW

Home visits by CHWs and educating families about danger signs in the postpartum period, examining newborns for danger signs, increasing referral compliance

1 month

Skills and knowledge on maternal and newborn care.

Yes

Not assessed

Not assessed

Not assessed

Not assessed

Assessed

Compliance with referral by the CHWs increased from 55.7% during the first three-month period of implementation to 80.1% during the third three-month period of implementation and was thereafter maintained at 75–80%.

There was a highly significant (p < 0.00001) increase in the proportion of families that sought care from qualified providers for sick newborns in the intervention arm and a non-significant increase in the comparison arm.

Broughton et al. 2016 [27]

Ecuador

Traditional birth attendant (TBA)

Provincial-level network to coordinate maternal newborn health services and strengthened linkages between levels of care

4 days

TBAs and CHWs were trained to identify maternal and newborn danger signs and risk factors and to refer them to health centers.

Yes

Not assessed

Assessed

Assessed

Not assessed

Not assessed

More mothers could identify newborn danger signs at endline (83%) compared to baseline (30%). Providers were able to identify two or more newborn danger signs 86% (n = 311) in first year of project vs. 98% (n = 49) in last year of project.

Darmstadt et al. 2011 [28]

Bangladesh

CHW

Home visits by CHWs to assess signs and symptoms of neonatal illness

36 days

Pregnancy surveillance, counseling and negotiation skills, essential newborn care, neonatal illness surveillance and management of illness based on a clinical algorithm adapted from IMCI.

Yes

Not assessed

Not assessed

Assessed

Not assessed

Not assessed

Compared to physician assessment, use of a 6-sign algorithm by CHWs had a sensitivity of 81.3% and specificity of 96.0% for identifying infant referral need and a sensitivity of 58.0% and specificity of 93.2% for screening infant mortality.

Darmstadt et al. 2010 [29]

Bangladesh

CHW

Preventive and curative maternal-neonatal healthcare package

36 days

As above.

Yes

Not assessed

Assessed

Not assessed

Not assessed

Assessed

Improvements in unprompted caregiver knowledge of maternal and neonatal danger signs were significantly larger in the intervention arm compared to the comparison arm. Among neonates who had ≥1 of the 10 selected complication signs, care seeking from a qualified provider increased significantly more in the intervention arm (from 31 to 56%) than in the comparison arm (from 27 to 35%).

Darmstadt et al. 2009 [30]

Bangladesh

CHW

Home visits by CHWs to assess signs and symptoms of neonatal illness

36 days

As above.

Yes

Not assessed

Not assessed

Assessed

Not assessed

Not assessed

Compared to physician assessment, CHWs classified very severe disease in infants with a sensitivity of 73%, specificity of 98%, PPV of 57% and NPV of 99%.

Das et al. 2014 [31]

India

Accredited Social Health Activists (ASHA)

ASHA home visits

5 days

Orientation on state government Home-based Newborn Care (HBNC) recording formats, and an Integrated Management of Newborn and Childhood Illness (IMNCI) skills review.

No

Not assessed

Not assessed

Assessed

Not assessed

Not assessed

Compared to trained investigators, ASHA agreement on the need to further assess infants was intermediate (kappa = 0.48, P < 0.001) and overall ASHA-investigator agreement on diagnosis was poor (kappa = 0.23, P = 0.01).

Findley et al. 2013a [32]

Nigeria

Control vs. Community volunteer (CV) vs. CHW and CV

High intensity (CV + CHW) compared to low intensity (CV) intervention

2 weeks

Community based using train-the-trainers model.

No

Not assessed

Assessed

Not assessed

Not assessed

Assessed

At follow-up, most women knew at least one of the newborn danger signs, with the most commonly known danger sign being high fever, known by 83–84% of women, regardless of the level of intervention they had experienced (F = 0.182, P = 0.825). However, at follow-up, for almost all danger signs, women living in the intervention communities were more aware than those in the control communities.

Findley et al. 2013b [33]

Nigeria

Control vs. CHW & CV

Health promotion through system-wide changes in health planning and implementation

Not specified

Community based using train-the-trainers model.

No

Not assessed

Assessed

Not assessed

Not assessed

Not assessed

At the midterm follow-up, most women knew at least one of the newborn danger signs, with the most commonly known danger sign being high fever, known by 82.7% in the control and 84.2% in the intervention communities. Many women knew other critical danger signs that indicated the need for the baby to be seen by a health worker. In the intervention areas, 31.0% knew to worry about diarrhea, dehydration, and sunken fontanel and about fitting or convulsions, significantly more than in the control areas. Women in the intervention areas were also more likely to know about breathing problems and not being able to suckle or refusing to feed.

Gebremedhin et al. 2019 [49]

Ethiopia

Health extension worker (HEW) and nurse

Use of HEWs at community levels to strengthen linkages between health centers and health posts

Not specified

Community-based newborn care.

No

Assessed

Assessed

Not assessed

Not assessed

Assessed

Provider knowledge of newborn related indicators ranged from poor to good.

88.8, 75.4 and 70.1% of mothers knew fever, poor sucking or ability to suck and fast breathing were danger signs.

Goel et al. 2019 [50]

India

ASHA

National government strategy of HBNC implemented by ASHAs as a quality of care improvement intervention package

2 days

Refreshing of technical skills, communication skills and counseling techniques; early initiation of breastfeeding, thermal care, hand washing, restrictive handling, cord care, recognition of danger signs.

No

Assessed

Not assessed

Not assessed

Assessed

Assessed

Knowledge score went from 5.1 to 7.1 (max score 9), with 39.2% relative change from baseline. There was an improvement in the intervention area for a mother recognizing illness in a neonate in the first month (mean 15.3 (SE 8.8–25.3) in standard area vs. 31.9 (SE 22.1–43.6) in the intervention area p = 0.019), and mother contacting an ASHA for neonatal illness (mean 2.8 (SE 0.8–9.6) in the standard area vs. mean 23.2 (SE 14.8–34.4) in the intervention area (p < 0.001).

Gupta et al. 2017 [34]

India

Auxiliary nurse midwife

Microteaching technique for enhancing the postnatal care skills of health workers

17 months

Training content was developed from the gaps identified while observing health workers delivering postnatal care in the videos.

Yes

Not assessed

Not assessed

Assessed

Not assessed

Not assessed

Out of a maximum score of 2, newborn examination of the eye (0.07), limping of legs (0.07), feeding (0.14), abdominal examination (0.0), and chest indrawing (0.0) were poor (average score < 0.30) at the baseline, but showed significant (p < 0.001) improvement ranging from 1.50 to 1.75 at the end of the final round. Average neonatal examination skill score improved from 0.52 (Round 1) to 1.29 (Round 2) to 1.74 (Round 3).

The overall skill assessment score significantly improved with each round of microteaching from 0.64 in the first round to 1.76 at the end of the third round.

Hodgins et al. 2010 [35]

Nepal

Female community health volunteer (FCHV)

Antenatal health education package

Not specified

District public health system supervisors and supervisors provided the bulk of the training and supervision.

No

Not assessed

Not assessed

Not assessed

Not assessed

Assessed

The percentage of respondents who sought care following recognition of danger signs increased for newborns: Baseline 51.0 (n = 310), Endline 59.4 (n = 414), OR: 1.42 (1.08–1.88). There was a decrease of about one half day in the time lag between illness onset and care seeking.

Horwood et al. 2017 [36]

South Africa

CHW supervisor

Quality improvement training for CHWs and supervisors

2 weeks

Community-based care of women and infants information on HIV and prevention of mother-to-child transmission (PMTCT), identification of signs of illness in newborn infants and children.

Yes

Not assessed

Assessed

Not assessed

Not assessed

Assessed

Mothers in the intervention group were more likely to have received a visit during pregnancy (75.7% vs. 29%, p < 0.0001) and postnatal period (72.6% vs. 30.3%, p < 0.0001). Mothers in the intervention group had higher maternal and child health knowledge scores (49 vs 43%, p = 0.007) and reported higher exclusive breastfeeding rates to 6 weeks (76.7 vs 65.1%, p = 0.02).

Khanal et al. 2011 [37]

Nepal

Facility-based community health worker (FB-CHW) & female community health volunteer (FCHV)

FCHV and FB-CHW use of MINI algorithm and subsequent administration of antibiotics

4 months

Assessment and management of neonatal infections and early newborn care (ENC) messages.

Yes

Not assessed

Not assessed

Assessed

Not assessed

Not assessed

FCHV assessments matched the more highly trained facility-based CHWs in over 90% of cases. Treatment was initiated in 90% of sepsis cases.

Kumar et al. 2008 [38]

India

CHW

Intervention comparing package of preventive essential newborn care with intervention group receiving essential newborn care plus liquid crystal sticker that indicates hypothermia by changing colour

7 days

Combination of classroom-based and apprenticeship-based field training on knowledge, attitudes, and practices related to essential newborn care within the community, behaviour change management, and trust-building.

No

Not assessed

Not assessed

Not assessed

Not assessed

Assessed

Improvements in birth preparedness, hygienic delivery, skin-to-skin care, umbilical cord care and breastfeeding were seen in intervention arms. Compared with controls, neonatal mortality rate was reduced by 54% in the essential newborn-care intervention (p < 0.0001) and by 52% in the essential newborn care plus ThermoSpot arm (p < 0·0001). There was little change in care-seeking.

Limaye et al. 2020 [51]

Bangladesh

Field worker (FW)

Educational digital health intervention providing netbook computers to FWs

2 days

eToolkits intended for use as a counseling tool and eLearning courses to supplement in-person trainings. Course included maternal and newborn health (MNH), family planning, nutrition, integrated messaging and interpersonal communication.

Yes

Assessed

Not assessed

Not assessed

Not assessed

Not assessed

Difference in mean scores in relation to newborn danger signs was significant from pre to post. Scores increased from 2.01 to 3.27 (range of 0–4) for a 1.25 mean score difference, significant at p < 0.001. Scores on how to care for pre-term infants increased from 0.56 to 1.49 (range 0–2) for a 0.93 difference, significant at p < 0.001.

Manu et al. 2016 [39]

Ghana

CBSV

Home visits in the first week of life; newborn care seeking promotion by CBSVs; dialogue and problem solving with families around barriers to seeking care

9 days

Interactive discussions, group exercises and practical newborn assessment video exercises.

Yes

Not assessed

Not assessed

Assessed

Not assessed

Assessed

Almost 70% of recently delivered women received CBSV assessments. Compliance with referrals was very high, especially in the poorest quintile. Independent care seeking for severe newborn illness increased from 55.4% in control to 77.3% in intervention zones. This was a near doubling among the poorest quintile.

Mascarenas et al. 2015 [40]

Kenya

CHW

Training program for CHWs to detect clinical signs that predict severe illness in children under two months of age

5 days plus 2 days on neonatal danger signs

Signs and symptoms of neonatal illness on a routine home visit during the first week of life, and determining need for referral to a health facility, methods of malaria control, childhood respiratory tract infections and diarrheal disease, domestic and personal hygiene, child nutrition, reproductive health, HIV/AIDS, and communication skills.

No

Not assessed

Not assessed

Not assessed

Not assessed

Assessed

35% of families who didn’t receive a CHW visit reported taking their infant to a healthcare facility, compared with 21% of families who did receive a CHW-visit (p < 0.01). Rates of overnight hospitalization were 6% for not-visited infants and 1% for visited infants (p < 0.01).

McConnell et al. 2016 [41]

Kenya

CHW

Phone calls using checklist by CHWs to newly delivered mothers

4 days

Conducting screenings using the checklist and counseling mothers and caregivers on essential postnatal health education.

No

Not assessed

Assessed

Assessed

Not assessed

Assessed

85% of women in the home visit arm were able to name 3+ maternal danger signs and 3+ infant danger signs. 81 and 78% in the phone call arm were able to name 3+ maternal danger signs and 3+ infant danger signs, respectively. 79 and 67% of women in the standard of care arm were able to name 3+ maternal and infant danger signs.

Facility-based care seeking for infants was high in all arms at 94–96%. However, the differences in timing of infant-related care seeking was statistically different between arms. Women in the phone call and home visit sought care for their baby 2.0 (p = 0.014) and 1.8 days (p = 0.034) earlier than the standard of care.

Mozumdar et al. 2018 [52]

India

Peer health educator

Home-based maternal and newborn care (HBMNC) messaging by self-help groups

1 day

Key HBMNC topics, conducting discussions on health topics and the different health services that are available at public health facilities.

Yes

Not assessed

Assessed

Not assessed

Not assessed

Not assessed

Women in the experimental area showed significant knowledge increase in importance of ANC, danger signs for newborn child, skin to skin care, KMC and delayed bathing. Endline results also showed significant increase in knowledge for cord care, delayed bathing and KMC. In the experimental arm, knowledge of danger signs for a newborn child all had significant increases, except for chest indrawing. For the comparison group, some danger signs showed a loss of knowledge between baseline and endline.

Nalwadda et al. 2013a [42]

Uganda

CHW

CHW visits

Not specified

Counseling for caretakers with referred newborns; newborn danger signs.

No

Not assessed

Not assessed

Not assessed

Not assessed

Assessed

700 newborns were referred by CHWs and successfully traced. Of these, 373 (53%) were referred for immunization and postnatal-care, and 327 (47%) were referred for a danger-sign. 439 (63%) complied, and of the 327 newborns with a danger sign, 243 (74%) caretakers complied with the referrals.

Nalwadda et al. 2013b [43]

Uganda

CHW

CHW home visit

5 days

CHWs’ roles during pregnancy and after delivery including health education, screening for danger signs and counseling for referral.

Yes

Assessed

Not assessed

Assessed

Not assessed

Not assessed

68% of the CHWs attained the pass mark for knowledge scores. 74% mentioned the required five newborn danger signs; chest in-drawing and grunting were never mentioned as newborn danger signs. 63% attained the pass mark for both skills and communication. 98% correctly identified the four case-vignettes as a sick or not sick newborn. Preterm birth was the least identified danger sign from the case-vignettes.

Namazzi et al. 2017 [44]

Uganda

CHW

Community mobilization & sensitization using CHWs and radio talk shows

5 days

Home visits and community dialogues on maternal and newborn issues.

Yes

Assessed

Not assessed

Not assessed

Not assessed

Not assessed

CHWs’ knowledge of maternal and newborn health improved from 41.3 to 77.4% after training, and to 79.9% one year after training. Main predictors of knowledge were age and level of education.

Ndaba et al. 2019 [53]

South Africa

CHW

Lectures, demonstrations and individual and group exercises; simulated home visits; training manual based on UNICEF and WHO materials focused on home visits, support and care, child feeding practices and postnatal care at a community level

3 days

Skills and knowledge to assist pregnant mothers and their babies pre and post-delivery at their homes in the first month of life; to conduct home visits during pregnancy and after delivery; and to enhance a good working relationship with the Community Advisory Group and Partner Defined Quality structures.

Yes

Assessed

Not assessed

Not assessed

Not assessed

Not assessed

CHWs’ knowledge of MNH improved from 63 to 72% (clinic A), 38 to 66% (clinic B), 48 to 50% (clinic C). The test focused on home care and support, child feeding and clinical care and support for MNH.

Nsibande et al. 2013 [45]

South Africa

CHW

CHW home visits

2 weeks

Role plays that were video recorded and content used for teaching and supervision purposes.

No

Not assessed

Not assessed

Not assessed

Not assessed

Assessed

95% of mothers completed the referral advised by CHWs, with 22% occurring during the first two weeks of life. For completed referrals, 51% of mothers could recognize danger signs that required care seeking (p = 0.01). The mothers who did not complete referral were unable to recognize infant danger signs.

Rahman et al. 2019 [54]

Bangladesh

CHW

CHW meetings with women and courtyard meetings with husbands and community members.

Not specified

Build capacities of women (promote birth preparedness and complication readiness, increase awareness of rights and MNH needs and increase health providers’ capacity to counsel women) and engage families and communities (male involvement and community involvement).

No

Not assessed

Assessed

Not assessed

Not assessed

Assessed

In the intervention area, women who were aware of at least 3 danger signs for newborns increased from 63 to 83% while knowledge declined in the comparison area. For all knowledge categories related to danger signs, the intervention had significantly increased levels. Regarding newborn care from a skilled health professional, there was no notable improvement between baseline and endline in either comparison or intervention areas.

Soofi et al. 2017 [46]

Pakistan

Lady health worker (LHW)

LHWs linked with traditional birth attendants and encouraged to attend home births.

5 days’ training for LHW programme master trainers; LHWs received initial 3 days of training and monthly 1 day sessions

High-risk pregnancies, antenatal care (ANC) care and other standard pregnancy advice, neonatal danger signs and illness with focus on neonatal sepsis; TBAs received basic essential neonatal care training and linkages with LHWs; support groups received training in communication, and counseling skills and knowledge on birth asphyxia, low birthweight, and sepsis; health-care providers received training on essential neonatal care and management of birth asphyxia, low-birthweight babies, and neonatal sepsis, provision of inflatable bag and mask and oral amoxicillin for sepsis with management protocols.

Yes

Not assessed

Not assessed

Not assessed

Not assessed

Assessed

For care seeking for neonates with reported illness, there was no significant difference between control or intervention (93 vs 93%, p = 0.89) However, there was a significant difference between control and intervention groups for neonates with possible infections who were seen by a LHW (2 vs. 29% (p < 0.001)).

Tripathy et al. 2016 [47]

India

ASHA

ASHA-led women’s group meetings

11 days

Not specified.

Yes

Not assessed

Not assessed

Not assessed

Not assessed

Assessed

Improvements in care seeking for newborn health problems were seen in both arms. From baseline to evaluation, the intervention arm improved from 46 to 72% and the control arm moved from 60 to 77%.

Waiswa et al. 2015 [48]

Uganda

CHW

Home visit package with health facility strengthening

5 days

Goal-oriented ANC, managing maternal complications, infection prevention, managing normal labour and partograph use, neonatal resuscitation, care of the sick newborn, and extra care for the small baby using kangaroo mother care.

Yes

Not assessed

Not assessed

Not assessed

Not assessed

Assessed

Improvements were seen in essential newborn care practices between intervention and control arms. Almost half (49.6%) of the mothers in the intervention waited more than 24 h to bathe the baby, compared to 35.5% in the control arm (p < 0.001) and dry umbilical cord care was also significantly higher in intervention areas (63.9% vs. 53.1%, p < 0.001).

Immediate and exclusive breastfeeding also saw improvements between intervention and control arms (72.6% vs. 66.0%; p = 0.016 and 81.8% vs. 75.9%, p = 0.042, respectively).

There was no difference in care-seeking for newborn illness in either arm but it was already high at 95%.