Pediatric Infectious Disease Journal:
Management of Newborn Infections in Primary Care Settings: A Review of the Evidence and Implications for Policy?
Bhutta, Zulfiqar A. MBBS, PhD*; Zaidi, Anita K. M. MBBS, SM*; Thaver, Durrane MBBS, MSc*; Humayun, Quratulain MBBS*; Ali, Samana FCPS*; Darmstadt, Gary L. MD, MS†
From the *Department of Paediatrics and Child Health, The Aga Khan University, Karachi, Pakistan; and †Department of International Health, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD.
Address for correspondence: Zulfiqar A. Bhutta, MB, BS, PhD, Department of Paediatrics and Child Health, The Aga Khan University, Karachi, Pakistan. E-mail: email@example.com.
Background: Long-term, sustainable programs to address high incidence and death rates from neonatal infections are required for improving child survival. There is an urgent need to define the role of community-based management for neonates with serious bacterial infections—-both at home and at first-level facilities.
Methods: We reviewed available evidence for community-based antibiotic management strategies for serious neonatal infections.
Results: Nine distinct studies contributing data for community-based management of neonatal pneumonia and sepsis were identified. In a pooled analysis of 5 controlled trials of community-based management of neonatal pneumonia (4 using cotrimoxazole, 1 ampicillin, or penicillin), all-cause neonatal mortality showed 27% [95% confidence interval (CI): 18%–35%] reduction and pneumonia-specific mortality, 42% (95% CI: 22%–57%). Substantial reductions in neonatal mortality have been demonstrated in a nonrandomized controlled study in rural India (62% reduction, P < 0.001) and in a cluster randomized trial in rural Bangladesh (34% reduction, 95% CI: 7%–53%). Reduced case fatalities (0%–3%) with community-based management of neonatal sepsis were observed in 2 small uncontrolled studies from India and Guatemala and a recent randomized trial from Pakistan.
Conclusions: Although methodological limitations preclude estimating the precise contribution of antibiotics toward neonatal mortality reduction in community settings in low income countries, available data suggest substantial benefit of case management approaches using antibiotics for neonatal sepsis in such settings.
Of the 4 million neonatal deaths in developing countries, a large proportion occur in domiciliary settings due to poor health service coverage, shortage of health care providers, and limited or lack of access to referral services and facilities.1,2 In addition, there is evidence to suggest that domiciliary care and care seeking practices may differ according to sociocultural norms3 and may also be influenced by gender issues.4 Appropriate strategies are needed for prevention of infections, as well as interventions for the domiciliary care and referral of newborn infants who develop bacterial infections. In several resource poor situations where prompt referral to a facility is not possible, health workers may have no alternative but to provide domiciliary care for the treatment of serious neonatal bacterial infections.3,5–7
Figure 1 indicates a conceptual model for the prevention and treatment of neonatal bacterial infections, given various risk factors. The model includes a range of preventive strategies that may act in synergy with therapeutic strategies to improve neonatal survival and can be incorporated together into packages of maternal and newborn care for potential scale up in primary care settings.8,9 Therapeutic strategies encompass both domiciliary care as well as care provided through first level community clinics and health outposts, frequently staffed by paramedical and nursing staff. The latter cadres may either work at these outposts or through outreach services delivering a mix of preventive and therapeutic interventions for maternal and newborn care.10
An in-depth evaluation of the evidence-base for preventive strategies is beyond the scope of this review, which focuses primarily on antibiotic management strategies for neonatal infections in developing country primary care settings. Recent studies of home treatment of sick newborns in rural India5 and Bangladesh7 have demonstrated significant reductions in neonatal mortality and have proven the feasibility of such approaches. This review examines the evidence base of home-based and first-level facility-based studies of treatment of neonatal bacterial infections. There are several context-specific assumptions on which community-based management of neonatal infections is being advocated (Table 1). Other issues such as criteria for selection of appropriate oral and/or parenteral antibiotics for use among neonates in appropriate settings, as well as pharmacological properties are explored in other articles of this series.11,12
We searched PubMed (last search in February 2008) for relevant studies conducted in countries classified as middle and low income economies using standardized search criteria for studies of community-based treatment of serious neonatal infections13 We reviewed all available observational reports, randomized controlled trials, systematic reviews, and meta-analyses, which included neonates and principally involved the management of serious neonatal infections in primary care (home or first-level facility) settings. Studies evaluating multiple interventions, which included community-based antibiotic case management for neonatal infections, were also reviewed. The available evidence was analyzed for neonatal mortality rates and case fatality with various regimens, antibiotic choice and duration, comparison data where available, delivery strategies, as well as methodological issues and other limitations. We excluded interventions that had principally been implemented in secondary or higher level facilities with little possibility of introduction in primary care settings.
Characteristics of Included Studies
Eleven distinct studies contributed data for neonates in this review6,7,14–22 plus an unpublished study (Zaidi et al, unpublished data 2007). These studies encompassed community-based management of pneumonia (2 studies, data in 4 published reports14,20–22) and sepsis (4 studies, data in 6 published reports5–7,15,16). Although studies of case management of pneumonia, which did not report disaggregated data for neonates were not included17–19,23–30; we included a meta-analysis31 of 5 trials of community-based management of pneumonia,14,17–20 which did obtain disaggregated data for neonates from 3 of the aforementioned excluded studies.14,17,19 Two studies contributing to the meta-analysis, which provided disaggregated data for neonates in their original report, have been discussed in greater depth in this review.14,20
Tables 2 and 3 describe characteristics of studies included, namely study location, design, number of neonates included, antibiotic regimens used, additional preventive or therapeutic interventions employed, delivery strategy, and outcomes such as neonatal mortality rate and infection-specific mortality rate in treated compared with untreated neonates, or baseline or comparison mortality rates, where available, for studies of primary diagnosis of neonatal pneumonia (Table 2) and neonatal sepsis (Table 3). All community-based studies reviewed (Tables 2 and 3) used clinical criteria for sepsis and/or pneumonia; none undertook laboratory investigations and hence did not contribute any bacterial etiology or antimicrobial susceptibility data.
Evidence for Community Management of Newborn Pneumonia
As highlighted in Table 2, the major body of evidence for the benefit of community-based case management of neonatal pneumonia is synthesized in the meta-analysis31 of 5 trials included conducted in India, Pakistan, Nepal, Tanzania, and Bangladesh14,17–20 which compared case management of neonatal pneumonia in community settings with a control group. Only 1 study, conducted in Tanzania, was reported as a randomized trial. Various antibiotics were used—most commonly oral cotrimoxazole.14,18–20 A variety of other interventions of varying intensity accompanied antibiotic case management in all intervention areas, including household visits, family mobilization, and immunization.31 All trials showed reduction in all-cause and cause-specific mortality in neonates, although this was statistically significant in only 2 trials, from India and Bangladesh.14,17 The pooled analysis showed a significant reduction in pneumonia-specific neonatal mortality of 42% [95% confidence interval (CI): 22%–57%; 4 trials],14,17,19,20 as well as a significant reduction in the overall neonatal mortality rate (NMR) of 27% (95% CI: 18%–35%; 5 trials).14,17–20
Pneumonia Case Management Trials Reporting Disaggregated Data for Neonates
A trial20 in a remote mountainous area of Nepal (which is included in the pneumonia case management meta-analysis),31 presented disaggregated data for neonates. It evaluated a community health worker-administered program of active case detection of childhood pneumonia (respiratory rate of ≥50/min or chest in-drawing) and home-based treatment with 5 days of oral cotrimoxazole. There were no facilities for referral and children who failed therapy were given oral chloramphenicol. Mothers were educated on signs of pneumonia and care seeking. Among neonates with early-onset pneumonia, there was a reduction in risk of death by 20% by the third year of intervention, although this was not statistically significant (0.80; 95% CI: 0.59–1.10).20 In further analysis, a statistically significant reduction, however, of 26% (0.74; 95% CI: 0.58–0.94) was found among infants aged 7 days to 5 months.
In the quasi-experimental pneumonia case management trial conducted by Bang et al14,21,22 in rural Gadchiroli—an underdeveloped rural district in Maharashtra, India—births and childhood deaths from 1988 to 1991 were prospectively recorded in 58 intervention and 44 adjacent control villages. Interventions included intensive training of paramedics, village health workers (VHWs), and traditional birth attendants (TBAs) in the examination of sick neonates and children, and recognition and treatment of pneumonia (presence of cough and a respiratory rate >50/min) with cotrimoxazole syrup and advice on continued breast-feeding. Referral was advised for very sick neonates, but if this was refused by families, they were managed at home with cotrimoxazole. In the combined analysis of 4620 neonates reported in 1994,22 pneumonia-specific mortality was significantly reduced by 11.7/1000 live births or 44.3% in the intervention area compared with the control area (P < 0.001). The overall mortality rate decreased by 15.9/1000 live births, or 20.2% (P < 0.01). Among 90 infants (1.95%) treated for pneumonia, the observed case fatality was 13.3%; it was 24% in the 0 to 7 day old infants treated, and 33% in those with referral indications.
Evidence for Community-Based Management of Neonatal Sepsis
Table 3 describes 4 published studies on community-based management of sepsis, including a nonrandomized quasi-experimental controlled trial,6 2 prospective observational studies,15,16 and 1 cluster randomized trial.7
Evidence From Observational Studies
Bartlett et al15 conducted a year-long, prospective, observational study on infant pneumonia and/or sepsis in a rural community in Guatemala. Three hundred twenty-nine infants surviving birth and the first day of life were followed for the first 3 months of life, biweekly home visits in the first month, and biweekly thereafter. Mothers were educated in detection of newborn illnesses. Thirty-four infants (10%) developed potentially life-threatening infectious diseases (92% of all serious medical problems encountered) and were treated in the community with various regimens (Table 3). The overall case fatality was 14%, and none of the babies managed wholly in the community died. An estimated NMR of 6/1000 live births was calculated using local population-based data, compared with a historical NMR of 39/1000 live births recorded for Guatemala, translating to an 85% reduction (P < 0.001) in NMR.
Bhandari et al16 observed 2007 infants aged 0 to 2 months at 2 first-level facilities (clinics) in urban slums in New Delhi over 12 consecutive months (Table 3). Infants were evaluated by trained pediatricians and hospital admission was recommended for 14% due to severe illness, among which 76% refused and were managed at home or as outpatients. A total of 124 patients with infectious illnesses were treated with oral cephalexin and intramuscular amikacin. The overall case fatality observed was 3.3%.
Evidence From Experimental Studies
In their landmark study,5,6,24 Bang et al evaluated a home-based package for neonatal care in rural Gadchiroli, India—an area with low female literacy and an extremely high baseline NMR of almost 100/1000 live births. Following baseline data collection (1993–1995), a comprehensive program for neonatal care was introduced in a stepwise manner in intervention villages (Table 3).
The first year of intervention (1995–1996) included visiting pregnant women in their homes, multiple postnatal visits for mother and neonate, training of TBAs in clean delivery, and training of female VHWs (1 per village) in the management of pneumonia (using cotrimoxazole) and other minor illnesses. In the following year, the VHWs were provided with basic resuscitation equipment and a neonatal care package (Table 3) and trained in recognition and home management of sepsis (collectively including sepsis, severe pneumonia, and/or meningitis). Referral was advised for babies with clinical sepsis (based on the presence on any 2 of 6 signs and in the event of refusal, VHWs administered home-based treatment using oral cotrimoxazole twice daily, and intramuscular gentamicin twice daily for 7 days in full-term babies and 10 days in preterm babies, with support for temperature maintenance and encouragement of breast-feeding, along with follow-up twice daily for 7 to 10 days. The package was continued relatively unchanged from 1998 to 20035,6,24 with the exception of removal of diarrhea and fast breathing alone from the clinical algorithm for case detection. Health outcomes were compared in 39 intervention and 47 control villages with no interventions. The results for each phase are presented in Table 3.
In the intervention area, neonatal mortality declined in year 1 by 21% compared with the control group, and 17% compared with the intervention area baseline rate, with the addition of household visits and home-based pneumonia case management. Following the introduction of the neonatal care package and intramuscular gentamicin for management of sepsis in year 2, the overall mortality declined by 28% compared with that observed in the control area, and by 29% compared with baseline. In year 3, subsequent to implementation of maternal health education, mortality rates declined further in the intervention group by 57.2% compared with the baseline control rates, and by 29% compared with the previous year in the intervention area. The overall decline in mortality from 62 to 25.5/1000 live births represented a 62% (P < 0.001) reduction relative to baseline control rates; sepsis-specific mortality was reduced by 76% (P < 0.005) from 27.5 to 6.6/1000 live births in the intervention area compared with the control.
The case fatality for suspected neonatal sepsis declined from 16.6% (year 1), before the introduction of the comprehensive care program, to 2.8% (years 2 and 3). Case fatality for neonates with sepsis among all cases treated by VHWs (during 1996–2003) was 6.9%, compared with 22% among untreated cases (probable cases of neonatal sepsis identified posthoc from verbal autopsies where the diagnosis was missed by VHWs, and those whose families refused treatment), a probable 68.6% reduction.6 The case fatality decreased in very low birth weight infants (<1500 g) from 69% in year 1, to 25% in year 3. Although a number of factors and interventions were operative during the entire intervention period (1996–2003), this concerted experience indicated that VHWs correctly diagnosed 492 cases of suspected neonatal sepsis (89%) and treated 470 neonates with antibiotics (8.9% of all neonates). Of 552 cases considered as sepsis by a computer algorithm, VHWs correctly treated 448 (81.2%) and gave unnecessary treatment to 22/470 (4.7%) of treated neonates. The case fatality was 6.9% in treated cases versus 22% in untreated or 16.6% in the preintervention period (P < 0.001). Altogether, home-based treatment resulted in a 67.2% reduction in case fatality among preterm neonates and a 72% reduction among low birth weight neonates, suggesting that this approach was feasible and effective.
In a recent cluster randomized trial from Sylhet in rural Bangladesh,7 24 clusters (with a population of about 20,000 each) were randomly assigned in equal numbers to 1 of 2 intervention arms (home care or community care) or to a comparison arm. Because of the study design, masking was not feasible. All married women of reproductive age (15–49 years) were eligible to participate. In the home-care arm, a nongovernmental organization partner, Shimantik, recruited 1 female community health worker (CHW) for every 4 villages (one per 4000 population) who identified pregnant women through routine surveillance during visits to each household once every 2 months, and made 2 antenatal home visits to promote birth and newborn-care preparedness. They made postnatal home visits to assess newborns on the first, third, and seventh days of birth, and classified sick neonates into 3 categories—very severe disease, possible very severe disease with more than one sign, or possible very severe disease with one sign. When CHWs diagnosed very severe disease or possible very severe, they referred neonates to subdistrict hospitals after giving an injectable dose each of procaine benzylpenicillin and gentamicin. If families were unable to go to a health facility but consented to home treatment, the CHWs continued treatment with procaine penicillin and gentamicin once daily for a total of 10 days. Neonates with possible very severe disease with 1 sign were not given antibiotics by the CHWs and were referred to subdistrict hospitals. When the family was not able to go to the hospital, community health workers made a follow-up visit to these neonates within the next 24 hours to monitor the infant for signs of illness and reinforce referral. In the home-care arm, female community mobilizers were recruited, each of whom took 8 months to visit her entire catchment area. In the community-care arm, each female community mobilizer was responsible for a population of 18,000, which was divided so that each geographical area of about 225 people was visited once every 4 months; male community mobilizers visited each area once every 10 months. Additionally, in the community-care arm, female volunteers called community resource people (usually TBAs from the same community) were recruited to identify pregnant women, encourage them to attend community meetings held by the community mobilizers, receive routine antenatal care, and seek care for signs of serious illness in mothers or newborns. Families in the comparison arm received the usual health services provided by the government, nongovernment organizations, and private providers.
In the last 6 months of the 30-month intervention, NMRs were 29.2 per 1000, 45.2 per 1000, and 43.5 per 1000 in the home-care, community-care, and comparison arms, respectively. Neonatal mortality was reduced in the home-care arm by 34% (adjusted relative risk 0.66; 95% CI: 0.47–0.93) during the last 6 months versus the comparison arm. No mortality reduction was noted in the community-care arm (0.95; 95% CI: 0.69–1.31).
Additional data available from a recent study of community management of suspected neonatal sepsis in Pakistan (Zaidi et al 2007, unpublished data) suggesting that CHWs can recognize serious neonatal infections and refer for treatment with injectable antibiotics in a primary care community health center. Among 434 newborn infants with suspected sepsis, an overall 86% success rate was seen for treatment with 1 of 3 regimens (daily injectable penicillin and gentamicin or daily injectable ceftriaxone or daily oral cotrimoxazole and injectable gentamicin). The latter regimen was found to be inferior to the penicillin and gentamicin combination and to the ceftriaxone regimen.
It is recognized that a range of preventive strategies can reduce the burden of neonatal infections in community settings and must be implemented at scale.3,5,9,10,32,33 While preventive strategies can be implemented in all cases and in at-risk populations, the key factor in treating neonatal infections successfully in community settings is the appropriate rapid diagnosis and triage to therapy. In contrast to infections in older infants and children, the onset of illness and course of progression is much more rapid in newborns, thus both clinical diagnosis as well as empirical therapy are the mainstay for management of neonatal sepsis.34 Figure 1 indicates a conceptual framework for considering preventive and therapeutic strategies for reducing the burden of neonatal infections in developing countries.
As our evidence review indicates, there is limited evidence for community-based management of serious neonatal infections. However, available data do suggest that case management approaches using antibiotics may play a substantial role in reducing mortality if integrated into home or community-based perinatal care packages.3,10,35
Several limitations must be recognized in evaluating these data. Three of the studies catering to newborn pneumonia.14,21,22 used the WHO guidelines (1989) or their 1991 revised version36 for pneumonia diagnosis and management. According to these WHO criteria, children with rapid breathing but no chest in-drawing (pneumonia) receive antibiotics as outpatients. Children who have chest in-drawing, with or without rapid breathing (severe pneumonia), are referred for hospitalization and treated with parenteral penicillin or ampicillin every 6 hours for at least 3 days, followed by oral ampicillin or amoxillin or intramuscular procain penicillin for at least 5 days. In the event of lack of recovery within 48 hours of initiation of therapy with penicillin, or deterioration, a change of antibiotic to chloramphenicol was recommended. Although these measures for the treatment of pneumonia have been modified in recent years, their application to the treatment of pneumonia in newborn infants has not been systematically evaluated.37,38 Most of the studies reviewed by Sazawal and Black31 are almost 2 decades old and some used a mix of measures such as active case detection, community mobilization, maternal health education and establishment of systems for referral by trained CHWs, and the criteria for triaging illness on the basis of disease severity may have changed over time. Notwithstanding the above, given the suggestion that a substantial part of neonatal sepsis may represent pneumonia,39 it is imperative that clinical algorithms for the recognition and management of neonatal sepsis and pneumonia are evaluated afresh for their validity and outcomes.
Evidence from trials of pneumonia case management among neonates, although limited by nonrandomized open comparisons, is nevertheless consistent in showing benefit of antibiotic case management approaches, largely using cotrimoxazole. Although the contribution of cointerventions, such as education and immunization, in addition to impact of methodological limitations, could not be quantified in the pooled analysis,31 they were estimated not to have major effects on the results of the meta-analysis. Indeed, the magnitude of reduction in neonatal mortality is comparable to the proportion of neonatal illness estimated to be due to infections,11 suggesting that substantial impact was achieved through case management. Notwithstanding the limitations and the relatively old data cited above, it is logical to assume that case management of pneumonia in community settings by health workers is feasible and may indeed lead to significant benefits.
Interpretations of the findings from the reported reduction in neonatal mortality from rural India22 are also limited by lack of a true control group and estimates of the impact of antibiotic management of sepsis derived from before-after comparisons of an observation phase (1995–1996) with an intervention phase (1996–2003) may be overestimated. In addition, estimates of impact of individual components obtained by “artificially disaggregating” an interdependent package of interventions, as the authors acknowledge, may exaggerate the benefit of each component.22 Recent studies on community case management of sick newborns do suggest that treatment is possible with antibiotics forming one part of a package of home care.
The large reductions in sepsis and pneumonia-specific mortality rates, as well as the very low case fatality rates demonstrated in community settings, suggest that home-based antibiotic management could be an attractive alternative to current recommendations of hospital admission and treatment with injectable antibiotics for resource poor countries with poorly functioning health care systems. There may however, be several impediments to widespread implementation of community-based programs for serious neonatal infections in resource poor countries. First, the cost and logistic considerations of such programs may be substantial, particularly those depending upon frequent household visits for case detection, management, supervision, and follow-up. For an estimated 83% of 26 million live births per year in India which occur at home,31 the cost of implementing home-based neonatal care in India alone is an estimated $1.14 billion per year (assuming $5.30 per neonate as documented in Gadchiroli with injectable antibiotics).14 Other recent studies from South Asia may provide a more accurate estimate of the cost effectiveness of interventions and preliminary data from studies done in Sylhet and Hala, Pakistan, do indicate that the costs of community-based intervention packages are substantially lower (Darmstadt et al, Bhutta et al; personal communication 2008).
Second, it must be emphasized that the available evidence on case management, particularly for neonatal pneumonia, largely predates the emergence of drug resistance among causative pathogens. Given recent trends in increasing antimicrobial resistance among common respiratory pathogens, it is imperative that antibiotic choices (especially the efficacy of oral cotrimoxazole) be validated in community settings.40 It is likely that oral amoxicillin, especially when given as a short course,41 may represent a better choice for the ambulatory management of nonsevere pneumonia in community settings. The need to monitor resistance to antibiotics among pathogens causing neonatal infections is 1 major concern in community-based antibiotic management strategies. Widespread antimicrobial resistance may preclude the ability to design simple, inexpensive, and effective antibiotic regimens in many countries. Neonatal pathogens also vary by region and temporally,42 which means that each country undertaking to develop and implement a case management program will need to periodically investigate the spectrum and antimicrobial resistance patterns of locally important pathogens in community-settings. This is particularly important in determining optimal antibiotic choices for empirical treatment of pneumonia.43 Recent data from a variety of settings do indicate that it may be possible to institute appropriate therapy guidelines at scale with reduction in rates of emergence of antibiotic resistant organisms.40
There are additional concerns such as the need for standardized diagnostic criteria, which are suitable for use by trained community health workers to detect sepsis within communities. Compared with computer-based diagnosis, community workers were able to correctly detect 89% of suspected neonatal sepsis, using criteria developed by Bang et al.5 The Young Infant Study group44 has recently developed an algorithm for predicting severe illness in neonates presenting to first-level facilities with reported sensitivities of 85% and 74%, and specificities of 75% and 79%, in 0–6 and 7–59 day age groups respectively or more sensitive but less specific than criteria used by Bang et al.45 In a community-based study in Bangladesh, CHWs were shown to have high validity compared with physicians (73% sensitivity, 98% specificity) when using an algorithm similar to the Young Infant Study algorithm during routine household surveillance, for identification of sick neonates needing referral to hospital.46 The clinical algorithm had good sensitivity and high specificity for identification of neonates needing referral (81% sensitivity, 96% specificity) and lower sensitivity for neonatal mortality (55% sensitivity, 93% specificity).47 It must be underscored however, that our current clinical methods for case detection of possible neonatal sepsis may be picking up a range of clinical illnesses other than bacterial infections, including viral infections. This is probable given the 70%–80% survival rates of newborn infants with suspected sepsis who did not receive antibiotics (Zaidi et al, personal communication 2008) and the impact on neonatal and young infant febrile illnesses observed in a recent study of maternal influenza vaccination in Bangladesh.48 Further validation studies of home-based algorithms are clearly needed in a variety of settings.
Even though studies have attempted syndromic evaluation of serious newborn infections,49 they do not concur on choice of antibiotic therapy. The selection of appropriate antibiotic regimens is impeded primarily due to insufficient data regarding etiological agents of serious neonatal infections in community settings in developing countries.39 Further issues in the selection of appropriate regimens, elaborated elsewhere11,12 include monitoring of potential adverse effects such as toxicity due to inappropriate dosing or altered gastrointestinal colonization, as well as use of unnecessary injections by community health workers which could exacerbate hepatitis B and C transmission, suitability of such regimens for various health workers, and acceptability in communities. With high levels of motivation and support, home-based care was found acceptable to communities in rural India—adjudged by high acceptance rate (91%) and no documented complications of injections such as infection or injury.5 Others have successfully demonstrated the benefit of community motivation and health worker training on improved referral rates and care seeking with the public and private sectors within a district health system,50 suggesting that perhaps a combination of home-based and facility-based strategies may be necessary in evolving health systems.
Although no direct comparisons exist of groups or communities receiving neonatal care packages with antibiotics versus neonatal care packages without antibiotics, studies have indicated that other simple interventions such as maternal education and training of birth attendants could also lead to declines in neonatal mortality rates in areas with very high burden of neonatal morbidity and mortality, and may—in contrast to antibiotic programs—be more sustainable in the long term.10 In an uncontrolled study in India, domiciliary management of high-risk neonates (such as preterm and low birth weight or those with feeding problems) by VHWs, led to a 25% decline in neonatal mortality in the absence of management of sepsis.51 It could also be deduced from the Gadchiroli trials that the impact of maternal education—which led to a 29% decline in neonatal mortality over the previous year—was comparable to the effect of adding intramuscular gentamicin and the neonatal care package.14 However, some of the further decline, which cannot be quantified, may have been due to maturation of the ongoing intervention. Other interventions which have been evaluated in more methodologically rigorous cluster randomized controlled trials, have shown considerable decreases in NMR of 30% (95% CI: 6%–47%) and 29% (95% CI: 17%–38%) respectively, with a maternal and newborn care program in Nepal that did not include antibiotics but instead used a participatory approach involving women's groups,32 training of TBAs, and use of clean delivery kits in rural Pakistan.52 Recent data from a community behavior change management strategy from Uttar Pradesh, India also indicate that major reductions in neonatal mortality are possible (51% reduction in NMR, 95% CI: 37%–62%) with promotion of behavior change targeted at multiple stakeholders, including the mother, family, and community providers of newborn care, and not including antibiotic use.32
In conclusion, available evidence suggests that antibiotics have a clear role in reducing neonatal mortality in low income areas and can be effectively administered in homes via trained health workers. However, issues surrounding sustainability and acceptability on the part of families, health care workers, and policy makers, as well as selection and implementation of packages of care integrated across the continuum of care—with appropriate preventive as well as therapeutic interventions for scaling up—are areas needing further development. In a recent evaluation of the evidence for integrated interventions for maternal, newborn and child survival in primary care settings,10 a clear case has been made for introducing packages of care for promotive, preventive, and therapeutic strategies in community settings and first-level facilities. These include the utilization of all available opportunities for treating sick newborn infants with infections, including home care where alternatives are not available.
1. Bhutta Z, Ali N, Hyder A, et al. Perinatal and Newborn Care in Pakistan: Seeing the Unseen! Karachi, Pakistan: Oxford University Press; 2004.
2. Lawn JE, Cousens S, Zupan J. 4 million neonatal deaths: When? Where? Why? Lancet. 2005;365:891–900.
3. Bhutta ZA, Ali S, Ali TM, et al. Strategies to reduce the burden of neonatal deaths in developing countries: pushing the envelope. Arch Dis Childhood. In press.
4. Willis JR, Kumar V, Mohanty S, et al. Gender differences in care-seeking for newborn infants in rural Uttar Pradesh, India. J Health Pop Nutr. In press.
5. Bang AT, Bang RA, Stoll BJ, et al. Is home-based diagnosis and treatment of neonatal sepsis feasible and effective? Seven years of intervention in the Gadchiroli field trial (1996 to 2003). J Perinatol. 2005;25(suppl 1):S62–S71.
6. Bang AT, Bang RA, Baitule SB, et al. Effect of home-based neonatal care and management of sepsis on neonatal mortality: field trial in rural India. Lancet. 1999;354:1955–1961.
7. Baqui AH, El-Arifeen S, Darmstadt GL, et al. Effect of community-based newborn-care intervention package implemented through two service-delivery strategies in Sylhet district, Bangladesh: a cluster-randomised controlled trial. Lancet. 2008;371:1936–1944.
8. Knippenberg R, Lawn JE, Darmstadt GL, et al. Systematic scaling up of neonatal care in countries. Lancet. 2005;365:1087–1098.
9. Bhutta ZA, Darmstadt GL, Hasan BS, et al. Community-based interventions for improving perinatal and neonatal health outcomes in developing countries: a review of the evidence. Pediatrics. 2005;115:519–617.
10. Bhutta ZA, Ali S, Cousens S, et al. Integrating maternal, newborn and child health: what difference can primary care strategies make? Lancet. In press.
11. Darmstadt G, Batra M, Zaidi A. Parenteral antibiotics for the treatment of serious neonatal bacterial Infections in developing country settings. Pediatr Infect Dis J. In press.
12. Darmstadt G, Batra M, Zaidi A. Oral antibiotics in the management of serious neonatal bacterial infections in developing country communities. Pediatr Infect Dis J. In press.
14. Bang AT, Bang RA, Tale O, et al. Reduction in pneumonia mortality and total childhood mortality by means of community-based intervention trial in Gadchiroli, India. Lancet. 1990;336:201–206.
15. Bartlett AV, Paz de Bocaletti ME, Bocaletti MA. Neonatal and early postneonatal morbidity and mortality in a rural Guatemalan community: the importance of infectious diseases and their management. Pediatr Infect Dis J. 1991;10:752–757.
16. Bhandari N, Bahl R, Bhatnagar V, et al. Treating sick young infants in urban slum setting. Lancet. 1996;347:1774–1775.
17. Fauveau V, Stewart MK, Chakraborty J, et al. Impact on mortality of a community-based programme to control acute lower respiratory tract infections. Bull World Health Organ. 1992;70:109–116.
18. Khan AJ, Khan JA, Akbar M, et al. Acute respiratory infections in children: a case management intervention in Abbottabad District, Pakistan. Bull World Health Organ. 1990;68:577–585.
19. Mtango FD, Neuvians D. Acute respiratory infections in children under five years. Control project in Bagamoyo District, Tanzania. Trans R Soc Trop Med Hyg. 1986;80:851–858.
20. Pandey MR, Daulaire NM, Starbuck ES, et al. Reduction in total under-five mortality in western Nepal through community-based antimicrobial treatment of pneumonia. Lancet. 1991;338:993–997.
21. Bang AT, Bang RA, Morankar VP, et al. Pneumonia in neonates: can it be managed in the community? Arch Dis Child. 1993;68:550–556.
22. Bang AT, Bang RA, Sontakke PG; for the SEARCH Team. Management of childhood pneumonia by traditional birth attendants. Bull World Health Organ. 1994;72:897–905.
23. Kielmann AA, Taylor CE, DeSweemer C, et al. The Narangwal experiment on interactions of nutrition and infections: II. Morbidity and mortality effects. Indian J Med Res. 1978;68(suppl):21–41.
24. Bang AT, Reddy HM, Deshmukh MD, et al. Neonatal and infant mortality in the ten years (1993 to 2003) of the Gadchiroli field trial: effect of home-based neonatal care. J Perinatol. 2005;25(suppl 1):S92–S107.
25. Datta N, Kumar V, Kumar L, et al. Application of case management to the control of acute respiratory infections in low-birth-weight infants: a feasibility study. Bull World Health Organ. 1987;65:77–82.
26. World Health Organization. Case management of acute respiratory infections in children: intervention studies. Report of a meeting. Geneva, Switzerland: World Health Organization; 1988.
27. Pandey MR, Sharma PR, Gubhaju BB, et al. Impact of a pilot acute respiratory infection (ARI) control programme in a rural community of the hill region of Nepal. Ann Trop Paediatr. 1989;9:212–220.
28. Roesin R, Sutanto A, Sastra K, et al. ARI intervention study in Kediri, Indonesia (a summary of study results). Bull Int Union Tuberc Lung Dis. 1990;65:23.
29. Sutrisna B, Frerichs RR, Reingold AL. Randomised, controlled trial of effectiveness of ampicillin in mild acute respiratory infections in Indonesian children. Lancet. 1991;338:471–474.
30. Lye MS, Nair RC, Choo KE, et al. Acute respiratory tract infection: a community-based intervention study in Malaysia. J Trop Pediatr. 1996;42:138–143.
31. Sazawal S, Black RE. Effect of pneumonia case management on mortality in neonates, infants, and preschool children: a meta-analysis of community-based trials. Lancet Infect Dis. 2003;3:547–556.
32. Kumar V, Mohanty S, Kumar A, et al. Impact of community-based behavior change management on neonatal mortality: a cluster-randomized, controlled trial in Shivgarh, Uttar Pradesh, India. Lancet. In press.
33. Darmstadt GL, Bhutta ZA, Cousens S, et al. Evidence-based, cost-effective interventions: how many newborn babies can we save? Lancet. 2005;365:977–988.
34. Vergnano S, Sharland M, Kazembe P, et al. Neonatal sepsis: an international perspective. Arch Dis Child Fetal Neonatal Ed. 2005;90:F220–F224.
35. Haws RA, Thomas AL, Bhutta ZA, et al. Impact of packaged interventions on neonatal health: a review of the evidence. Health Policy Plan. 2007;22:193–215.
36. Technical basis for the WHO recommendations on the management of pneumonia in children at first-level facilities. Geneva, Switzerland: World Health Organization; 1991.
37. Bhutta ZA. Dealing with childhood pneumonia in developing countries: how can we make a difference? Arch Dis Childhood. 2007;92:286–288.
38. Bhutta ZA. Managing severe pneumonia in children in developing countries. Bmj. 2008;336:57–58.
39. Wardlaw T, Salama P, Johansson EW, et al. Pneumonia: the leading killer of children. Lancet. 2006;368:1048–1050.
40. Bhutta ZA. Drug resistant infections in poor countries: a major burden on children. BMJ. 2008;336:948–949.
41. Haider BA, Saeed MA, Bhutta ZA. Short-course versus long-course antibiotic therapy for non-severe community-acquired pneumonia in children aged 2 months to 59 months. Cochrane Database Syst Rev. 2008:CD005976.
42. Newton O, English M. Young infant sepsis: aetiology, antibiotic susceptibility and clinical signs. Trans R Soc Trop Med Hyg. 2007;101:959–966.
43. Bhutta ZA. Treating childhood pneumonia empirically: several issue still remain! Indian Pediatrics. 2008;45:538–539.
44. Young Infants Clinical Signs Study Group. Clinical signs that predict severe illness in children under age 2 months: a multicentre study. Lancet. 2008;371:135–142.
45. Bang AT, Bang RA, Reddy MH, et al. Simple clinical criteria to identify sepsis or pneumonia in neonates in the community needing treatment or referral. Pediatr Infect Dis J. 2005;24:335–341.
46. Darmstadt GL, Baqui AH, Choi Y, et al. Validation of community health workers’ assessment of neonatal illness in rural Bangladesh. Bull World Health Organ. In press.
47. Darmstadt GL, Baqui AH, Choi Y, et al. Validation of clinical algorithm to identify neonates with severe illness during routine household visits in rural Bangladesh. Bull World Health Organ. In press.
48. Zaman K, Roy E, Arifeen SE, et al. Effectiveness of maternal influenza immunization in mothers and infants. N Engl J Med. 2008;359:1555–1564.
49. Weber MW, Carlin JB, Gatchalian S, et al. Predictors of neonatal sepsis in developing countries. Pediatr Infect Dis J. 2003;22:711–717.
50. Bhutta ZA, Memon ZA, Soofi S, et al. Implementing community-based perinatal care: results from a pilot study in rural Pakistan. Bull World Health Organ. 2008;86:452–459.
51. Pratinidhi A, Shah U, Shrotri A, et al. Risk-approach strategy in neonatal care. Bull World Health Organ. 1986;64:291–297.
52. Jokhio AH, Winter HR, Cheng KK. An intervention involving traditional birth attendants and perinatal and maternal mortality in Pakistan. N Engl J Med. 2005;352:2091–2099.
neonatal; sepsis; pneumonia; case management; community
© 2009 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.