Neonatal health: Epidemiology, interventions, key challenges

1
Neonatal health Epidemiology, interventions,
key challenges

• Gary L. Darmstadt, MD, MS
• Director, International Center for Advancing
  Neonatal Health
• Department of International Health
• Bloomberg School of Public Health
• Johns Hopkins University

2
Outline

• Burden of neonatal mortality
• Effective interventions
• Preventive, family-community care
• Infection prevention and management
• Implementation of a package of essential newborn
  care, including sepsis/pneumonia management
• Major challenges/gaps

3
Based on Lancet 20053651147-52
20063671487-94
4
Causes of 4 Million Newborn Deaths
LBW
1.44 million
0.94 million
Disabilities/ Impairments??
Fresh stillbirths 1.3 million
Disabilities/ Impairments??
1.1 million
Disabilities/Impairments??
Lawn JE et al, Lancet 2005365891-900
5
Half of neonatal deaths are in the first 24 h
links with maternal health care are critical
Timing of 4 million newborn deaths
75 of neonatal deaths are in the first week
3 million deaths
6
Timing of cause-specific neonatal mortality
Uttar Pradesh, India
Sepsis/pneumonia
Baqui AH et al, Bull WHO 200684706-13
7
Interventions evidence of efficacy for
prevention or treatment of neonatal infections
Universal
Situational
Additional
Lancet 2005365977-88
8
Up to 2.5 million babies a year could be saved
yet these interventions do not reach those
in greatest need

• Up to 37 of neonatal deaths could be averted
  with family and community -based interventions,
  feasible now
• To reach the MDG targets, skilled care at
  facility level will also need to be scaled up

9

Shivgarh, Uttar Pradesh, IndiaEssential
preventive newborn care promoted through behavior
change communications (family package)

• Birth preparedness
• Clean delivery, clean cord and skin care
• Thermal care Immediate wiping, drying and
  keeping the baby warm
• Breastfeeding promotion
• Skin-to-skin care

Based on formative research on beliefs,
practices, roles
10
Intervention Strategy
Follow-up
Household
Household Neighbors
Community
DAY 1/2, 7, 28
Visits D 0/1 3/4
Visit II
Visit I
Mobilization Social mapping Community meetings
Antenatal Period
Postnatal Period
Independent Evaluation Team
SAKSHAM KARTA
Intervention by
Community Mobilizer Saksham Karta
9
Cluster-randomized controlled trial in 104,000 in
Shivgarh Block, Rae Bareli
11
Shivgarh Changes in Key Practices Acceptance of
skin to skin care
BCC
BCC Thermospot
Change in behavior depends on a visit
Comparison
12
(No Transcript)
13
Shivgarh Changes in Key PracticesBreastfeeding
on day 0
Some change in practice without a visit
14
Shivgarh Impact on Neonatal Mortality
Intervention arms
Comparison

• 40 reduction in perinatal mortality
• 50 reduction in neonatal mortality

15
Prevention of Infection
16
Skin and cord cleansing trial (Sarlahi, Nepal)
Design
17
Skin and cord cleansing Interventions

• Skin Cleansing Trial
• Pampers baby wipes (Procter Gamble Co.)
• Standard product (placebo)
• Standard product plus 0.25 chx

• Cord Care Trial
• Education of the family (dry cord care alone)
• Education soap water cleansing of the cord
  (Ivory LiquiGel) on days 1,2,3,4,6,8,10
• Education 4 chlorhexidine cleansing of the
  cord on days 1,2,3,4,6,8,10

18
Risk factors for umbilical cord infections in
Sarlahi, Nepal (n15,092)
Breast milk, saliva, water, other oils, herbs,
spices, curry, and others
19
Newborn Skin Cleansing
NBW infants
Chlorhexidine
LBW infants
Placebo
20
Skin cleansing Post vs Pre-randomization
mortality
21
Photo of umbilical cord showing extensive redness
around the base of the cord stump (approximately
72 hours after birth)
22
Chlorhexidine Cord Cleansing Impact on cord
infections in Nepal1
1Soap-water treatment had no effect. 2Mullany LC
et al. Arch Dis Child Fetal Neonatal Ed
200691F99-F104. 3Incidence density expressed as
cases per 100 child months, or neonatal
periods. 4Comparison between Chx and dry cord
care.
23
Chlorhexidine Cord CleansingImpact of early
intervention on mortality
Chlorhexidine group compared to dry cord care no
impact among soap-water group, no effect
modification.
Mullany et al. Lancet 2006 367910-918.
24
Recognition of Infection
25
Multicentre Study of Clinical Signs Predicting
Severe Illness in Young Infants(Young Infant
Clinical Signs Study 2)

• 6 countries, 9 sites
• To develop an evidence-based IMCI algorithm for
  sick young neonates in the first week of life
• To validate and improve the existing algorithm
  for infants aged 7-59 days.
• Existing IMCI algorithm 14 signs

26
Study design

• Conducted at health facilities to enable gold
  standard assessments of illness (including
  laboratory investigations) to be made
• Patients self-referred (no prior contact with a
  qualified provider)
• Person A representative of a first-line
  facility-based health worker
• Recorded historical factors, symptoms and
  presence of clinical signs
• Person B pediatrician who made gold standard
  assessment of whether or not the infant was ill
  enough to need urgent care a health facility

27
Young Infant Clinical Signs Study 2
28
Performance of algorithms using data from Young
Infant Clinical Signs 2 Study
Sensitivity
0-6 d
7-59 d
Specificity
0-6 d
7-59 d
29
Young Infant Clinical Signs Study 2Conclusions

• 8-sign algorithm had good sensitivity (87 in 0-6
  day, 78 in 7-59 day age group), but lower
  specificity (66 and 72, respectively).
• A single algorithm with 8 clinical signs may be
  recommended for identifying young infants from
  birth to 2 months who require hospital admission,
  among those brought to health facilities.
• New algorithm is simpler, has higher specificity.
  
• Further research is needed to develop and
  evaluate algorithms for screening newborns for
  illness in the community, such as during routine
  home visits.

30
Home- and Community-based Management of Newborn
Infections Lessons from Sylhet District,
Bangladesh
31
Projahnmo 2 (Sylhet) Background and Objectives

• Study Site 3 sub-districts of Sylhet district,
  Bangladesh, estimated population 500,000
• 13 skilled attendance at delivery
• NMR in trial area 50 / 1,000 live births
• 47 of neonatal deaths due to infections
• Evaluated the impact of a package of maternal and
  neonatal care interventions using two service
  delivery strategies
• Home-care
• Community-care

32
Prohahnmo 2 (Sylhet) Description

• Projahnmo intervention was implemented using
  MOHFW and NGO infrastructures
• CHWs and CMs were recruited through an NGO
  partner Shimantik CHWs trained for 6 weeks
• MOHFW facilities used for referral level care
• Community care Use of existing facilities
• In Home Care model, CHWs provided health
  education through 2 ANC visits, assessed newborns
  (days 1, 3/4, 7) using an IMCI algorithm adapted
  for use in surveillance in the community
• CHW performance was validated
• Referred sick newborns to sub-district hospitals
• Treated sepsis at home using injectable
  antibiotics if referral failed

33
Clinical Algorithm
34
Incidence of VSD and PVSD 2004-2005 (n8,474)
Number
Percent
Case fatality rate (CFR)
Morbidity
Very Severe Disease (VSD)
5.6
13.4
478
Possible VSD multiple signs (PVSD-MS)
131
1.5
8.4
Possible VSD single sign (PVSD-SS)
820
9.7
1.0
35
Timing of identification of VSD and PVSD
Home visits
36
Referral compliance and management of
sick newborns
Referral successful n ()
Treated by CHWs n ()
Treated by other providers n ()
No outside care sought n ()
Classification
162 (34)
204 (43)
25 (5)
87 (18)
VSD
32 (25)
49 (37)
12 (9)
38 (29)
PVSD-MS
79 (10)
522 (64)
35 (4)
184 (22)
PVSD-SS
Other includes village doctor, pharmacist,
homeopath
37
Case fatality rate by type of management
Referral successful d/n ()
Treated by CHWs d/n ()
Treated by other providers d/n ()
No outside care sought d/n ()
Classification
23/162 (14.2)
9/204 (4.4)
8/25 (32.0)
24/87 (27.6)
VSD
1/32 (3.1)
0/49 (0.0)
1/12 (8.3)
9/38 (23.7)
PVSD-MS
0/79 (0.0)
2/522 (0.4)
2/35 (5.7)
4/184 (2.2)
PVSD-SS
Other includes village doctor, pharmacist,
homeopath d of death, n in category
38
(No Transcript)
39
Projahnmo revised clinical algorithm for use in
household surveillance Very severe disease

• Not able to feed (based on feeding assessment)
• Convulsion observed
• Unconscious
• Lethargic
• Movement only when stimulated
• Fast breathing ( 60/min or more)
• Severe chest in-drawing present
• Weak, abnormal or absent cry
• Temperature gt 101ºF or lt 95.5ºF
• Umbilical pus and redness of the cord extending gt
  2 cm onto the abdomen

40
Conclusions and RecommendationsIs home-based
management feasible and effective?

• Feasible CHWs were able to use the Clinical
  Algorithm to assess, identify, and manage
  neonates in community with potentially serious
  illnesses
• Safe and effective
• Community accepted CHWs giving injectable
  antibiotics
• Recognition of newborn illnesses largely remained
  dependent on CHW home visits

41
Home- or community-based management of newborn
infectionsRecommendation for programs

• In settings where the health system is weak and
  care seeking is low (e.g., Sylhet),
  family-community care should be introduced
  universally and a phased introduction of home- or
  community-based sepsis management should be
  considered

42
Home-based management of newborn infections
what is next?

• All newborn infections do not need injectable
  antibiotics perhaps only VSD and PVSD-MS cases
  (7-8)
• Remaining cases perhaps may be effectively
  treated using oral antibiotics or no antibiotics
• Need to develop simplified treatment regimens,
  e.g., shorter course injectable therapy?
• Need to develop and test alternative, simplified
  delivery strategies (e.g., Uniject-gentamicin)

43
Major challenges

• Reduce deaths due to infections substantial
  progress can be made now
• Links with maternal health care
• Birth asphyxia identification and management
• Stillbirths
• Urban models
• Child development