Impact of 4.0% chlorhexidine cleansing of the umbilical cord on mortality and omphalitis among newborns of Sylhet, Bangladesh: design of a community-based cluster randomized trial
© Mullany et al; licensee BioMed Central Ltd. 2009
Received: 2 September 2009
Accepted: 21 October 2009
Published: 21 October 2009
The World Health Organization recommends dry cord care for newborns but this recommendation may not be optimal in low resource settings where most births take place in an unclean environment and infections account for up to half of neonatal deaths. A previous trial in Nepal indicated that umbilical cord cleansing with 4.0% chlorhexidine could substantially reduce mortality and omphalitis risk, but policy changes await additional community-based data.
The Projahnmo Chlorhexidine study was a three-year, cluster-randomized, community-based trial to assess the impact of three cord care regimens on neonatal mortality and omphalitis. Women were recruited mid-pregnancy, received a basic package of maternal and neonatal health promotion messages, and were followed to pregnancy outcome. Newborns were visited at home by local village-based workers whose areas were randomized to either 1) single- or 2) 7-day cord cleansing with 4.0% chlorhexidine, or 3) promotion of dry cord care as recommended by WHO. All mothers received basic messages regarding hand-washing, clean cord cutting, and avoidance of harmful home-base applications to the cord. Death within 28 days and omphalitis were the primary outcomes; these were monitored directly through home visits by community health workers on days 1, 3, 6, 9, 15, and 28 after birth.
Due to report in early 2010, the Projahnmo Chlorhexidine Study examines the impact of multiple or single chlorhexidine cleansing of the cord on neonatal mortality and omphalitis among newborns of rural Sylhet District, Bangladesh. The results of this trial will be interpreted in conjunction with a similarly designed trial previously conducted in Nepal, and will have implications for policy guidelines for optimal cord care of newborns in low resource settings in Asia.
Of the annual 3.7 million global neonatal deaths, almost all (99%) occur in developing countries, and one-third to one half are attributed to infections such as sepsis, meningitis, pneumonia, omphalitis, tetanus, and diarrhea. In settings where home delivery is common and attendance by skilled personnel is low, many babies are born in unhygienic conditions and infections of the umbilical cord stump are common [2, 3]. Exposure of the freshly cut umbilical cord stump to pathogens on the cutting instrument, on the hands of caretakers, or elsewhere in the environment leads to local cord infections (omphalitis) that may progress to systemic infection and death [2–4]. Alternatively, during the period immediately following birth, while the vessels of the umbilical cord remain patent, direct exposure to the bloodstream via the cord may lead to systemic infection without any visible signs of local cord infection .
The World Health Organization (WHO) currently recommends keeping the cord clean and dry. The guidelines do not include a universal recommendation for application of any topical antiseptic, although their potential role is acknowledged for settings where harmful practices are common . Practical and effective implementation of this guidance, however, is challenging in both hospital and community-based low resource settings and is often less than optimal. Recently, there has been increased interest in both the overall role of exposure of the cord stump to invasive pathogens and the potential benefit of topical cord cleansing with chlorhexidine, a safe and effective broad-spectrum antiseptic.
A large community-based trial in rural southern Nepal conducted between 2002 and 2006 randomized babies within clusters to receive one of three cord care regimens: (1) 4.0% chlorhexidine cleansing for 7 of the first 10 days after birth or (2) soap and water cleansing for 7 of the first 10 days after birth, or (3) dry cord care. Mothers of all infants received educational messages about hygienic cord care (e.g., hand-washing, clean cutting, and avoidance of harmful topical applications). Depending on the severity of infection, omphalitis incidence was reduced by 32-75% among infants receiving chlorhexidine compared to those in the dry cord care group. Overall, mortality among enrolled infants was 24% lower in the chlorhexidine group compared to dry cord care. About two-thirds of the infants were reached within 24 hours of birth and evidence of a protective benefit of chlorhexidine cleansing among this subset was increased. Chlorhexidine cleansing reduced severe infection by 87% and mortality by 34% among those enrolled within 24 hours, while no difference between the groups was observed when cord cleansing was initiated after 24 hours.
An expert consultative meeting convened in Washington, DC shortly after completion of the Nepal trial concluded that, prior to any changes in policy, at least one replication trial in South Asia would be necessary to expand the evidence base for chlorhexidine cleansing of the cord. Two specific questions were noted. First, would a similar beneficial effect of multiple cleansings of the cord through the first week of life be observed in a separate but similar population? Second, can a simpler regimen of cord cleansing only once and as soon as possible after birth, provide a similar protective benefit against omphalitis and mortality? Answering this second question is necessary as the intensive intervention (cleansing on 7 of first 10 days of life) followed in the Nepal efficacy trial might not be easily implemented in most programmatic settings, given associated logistical and financial barriers for outreach health workers. Furthermore, even if the intervention was delivered by caretakers themselves, compliance with the promoted instructions might be increased if the more simple recommendation of one-time cleansing as soon as possible after birth could be given. Moreover, a single application of antiseptic to the cord could more readily be incorporated into clean birth kits.
Thus, in addition to replicating the question of efficacy of multiple (i.e. daily for 7 days) cord cleansing with chlorhexidine (4%, as used in Nepal), it was deemed essential to also investigate the potential efficacy of a single application as soon as possible after birth. The second major research activity for the Projahnmo study site in rural Sylhet District of north eastern Bangladesh was designed to answer these questions and provide evidence to inform global guidelines for optimal umbilical cord care of newborns. In this manuscript, we discuss in detail the design and implementation of this community-based cluster randomized trial, including study site, randomization and intervention delivery procedures, follow-up data collection, and analysis plans.
Trial Design and Preparation
This community-based trial in Sylhet, Bangladesh was designed to assess the impact of chlorhexidine cleansing on neonatal mortality and omphalitis. Specifically, these outcomes were compared among babies randomly assigned within clusters to one of three cord care regimens: 1) Multiple Chlorhexidine Cleansing Group: Neonates received cleansing of the cord with 4.0% chlorhexidine as soon as possible after birth and on subsequent days until age 7 days; 2) Single Chlorhexidine Cleansing Group: Neonates received a single cleansing of the cord with 4.0% chlorhexidine as soon as possible after birth. 3) Dry Cord Care Group: Neonates in this group received no specific umbilical cord care ("dry cord care", as recommended by WHO) beyond the basic messages related to clean cord cutting and avoiding home-based applications to the cord, which were directed to all groups. Mothers and newborns in all three groups also received a basic community-based package of essential newborn care (described below).
This study was conducted in three sub-districts (upazillas) of Sylhet District (Zakiganj, Khanaighat, Beanibazar) of north eastern Bangladesh. This is the research site of Projahnmo, a collaboration of the Johns Hopkins Bloomberg School of Public Health, the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B), the Ministry of Health of the Government of Bangladesh, and Shimantik http://www.shimantik.org, a non-governmental organization. In the first Projahnmo study, details of which have been previously published, 24 unions (a government-defined administrative sub-unit of the upazilla, of approximately 20,000 population) in this region were randomized to one of three arms. Two arms received a newborn care intervention, delivered either through community-mobilization activities or a more intensive home-care approach (pregnancy and postnatal visits), while the third arm received currently available care, and served as the comparison arm. In the home-care arm, community-based health workers (CHWs) were additionally trained to assess newborns for severe illnesses and refer or treat sick newborns. In that trial, neonatal mortality was reduced by 34% in the home-care arm relative to the comparison arm. This study site and the home-care intervention including essential newborn care promotion and follow-up assessments of newborns formed the basis of the infrastructure and service provision upon which the Projahnmo Chlorhexidine Study was implemented.
Study Population and Surveillance Infrastructure
Each defined CHW area (cluster) was further divided based on population size into 4-5 village-based sub-sections of approximately 1000 population (~200 households) each. Within each cluster, a locally resident female CHW was recruited and trained in all aspects of the study. Similarly, within each of the cluster sub-sections (n = 590), a single female village-based health worker (VHW) was identified to assist directly with delivery of the cord care intervention.
Characteristics of the clusters participating in the Projahnmo Chlorhexidine Study and distribution across study arm and previous project status
Dry Cord Care Arm
Previous Projahnmo Status
- Home Care Arm
- Community Care Arm
- Comparison Arm
- Expanded Areas
Total Population Size
Mean Population Size
Number of VHW areas
Preparation of Chlorhexidine
At regular intervals, 20% chlorhexidine digluconate was purchased in bulk from ACI Limited, Bangladesh and transferred to Sylhet in 5 litre containers. The source manufacturer for ACI was Vipor Chemicals PVT. LTD http://www.viporchem.com based in Baroda, India. A Projahnmo worker prepared the diluted 4.0% chlorhexidine solution for use in the trial, by mixing distilled water (645 ml/litre) with the stock solution (355 ml/litre) to the appropriate concentration (7.1% of the total digluconate salt); this was identical to the dilution process in the previous Nepal trial. The diluted solution was individually packaged into 70 ml and 10 ml opaque bottles for distribution to VHWs in the multiple and single cleansing groups, respectively. For each batch of prepared solutions, 5 bottles were randomly selected, transferred to a collaborating pharmaceutical laboratory at Eskayef Bangladesh Limited, and analyzed for final CHX concentration.
Enrolment of Pregnant Women and Basic Intervention Package
Identification and recruitment of pregnant women was based on regular monitoring by CHWs of printed registers containing the names and locations all married women of reproductive age (MWRA) within their individual target areas. Workers updated the list of MWRA and ascertained the pregnancy status of all women on the register by conducting an informal survey of each household in their area once every 8 weeks. The study was explained, participation offered, and informed consent provided to each newly pregnant women identified through this monitoring process. Those agreeing to participate provided data including age, parity, date of last menstrual period, occupation, literacy, abbreviated birth history, and socio-economic information about the household.
Participating women were provided with a basic package of maternal and newborn health promotion messages and interventions, delivered at two home visits during pregnancy. The first of these sessions was conducted at the same time as the recruitment visit at about 3-4 months of pregnancy; the second occurred at approximately the 8th month of pregnancy. At each session, messages and counselling were provided regarding pregnancy care (antenatal care-seeking, tetanus toxoid immunization, iron/folic acid supplementation, and recognition of danger signs during pregnancy and at the time of delivery), identification and selection of a birth attendant, clean and hygienic delivery, early and exclusive breastfeeding, thermal care of the newborn, and postnatal danger signs and care-seeking. All families were provided a clean birthing kit and the use of items in the kit was explained through demonstration at these pregnancy visits.
Sessions during pregnancy provided opportunities to establish the process for notification of project workers regarding pregnancy outcomes, and to reinforce the importance of accomplishing this as rapidly as possible. By accompanying the CHWs on their two pregnancy home visits, the VHWs were able to orient the family members on the process of sending a message to the VHW via a family member or another individual when the pregnant women began labor. Compliance with this notification system was strengthened by explaining that rapid notification would enable faster assessment of the health of the baby by the CHW.
After being notified of an ongoing labor or recent delivery, the local VHW travelled to the home of the newborn to deliver the assigned intervention. Depending on the cord care regimen randomly allocated to that location (i.e. cluster), the VHW either provided the standard set of dry cord care promotional messages (dry cord care group) or applied the initial cleansing of the cord (single and multiple cleansing groups). VHWs in all groups visited the home of the newborn daily through the first week of life (i.e. 7 days) and basic messages about cord hygiene were promoted across all the groups at these visits. In addition, the VHW provided the chlorhexidine cleansing intervention at each visit in the multiple-cleansing arm, and at the first visit only in the single-cleansing arm. Each individual VHW consistently followed only one set of procedures throughout the trial period, as the unit of randomization was the area (cluster) of the VHW's supervising CHW. The VHW recorded the date and approximate time of visits, the vital status of the baby (alive/dead), and the status of the intervention for that day (provided, refused) on a pictorial form.
For newborns receiving chlorhexidine interventions (i.e. arms 1 and 2), the basic cleansing procedure followed by the VHW on day 1 was consistent with how the intervention was delivered in the previous community-based trial in rural Nepal. First, the VHW washed her hands with soap and water and allowed them to air dry. If the baby was wrapped or clothed, the VHW removed an amount of covering necessary to expose the cord stump area. She moistened a cotton swab by placing it on the mouth of the chlorhexidine bottle (10 ml or 70 ml bottle, depending on study allocation) and inverting the bottle. The worker then gently dabbed the abdominal skin around the base and cord stump thoroughly, ensuring that the entire area was covered with solution. A second cotton swab was similarly moistened and used to cleanse the tip of the cord (if not yet separated) or the middle of the cord stump (if separation had occurred). After cleansing the cord, the VHW wrapped the baby warmly and placed the baby in the arms of the mother or other family member. Emphasis was placed on thoroughly completing the cleansing procedure as quickly as possibly, followed by prompt wrapping of the baby.
In all clusters, after conducting the first visit after birth, the VHW notified their supervising CHW through direct communication, mobile phone, or secondary messenger. The supervising CHW then visited the home to initiate the neonatal follow-up and assessment schedule. At the first follow-up visit by the CHW (day 1), basic information regarding the characteristics of labor and delivery and morbidity prior to, during, and after delivery was recorded. Newborn-specific data collected at this visit included date and time of birth, sex, weight of the infant, and care given during and immediately after birth (bathing, massage, breastfeeding practices). At this first visit and subsequent assessment visits (days 3, 6, 9, 15) the CHW assessed the status of the newborn through a combination of direct observation (vital status, axillary temperature, chest indrawing, umbilical and skin infection, jaundice, lethargy, respiratory rate, etc) and questions directed to the mother regarding signs of morbidity of the infant and care-seeking since the prior visit (or since birth).
A particular focus was placed on assessing signs of umbilical cord infection. Previous community-based work in Nepal and Tanzania utilized a standard set of fifty digital images of umbilical cords with varying degrees of inflammation to assess intra- and inter-worker agreement in recognition of signs and to estimate sensitivity and specificity of workers compared to a physician gold standard ranking of the images. Following this model, CHWs were trained to recognize and classify the severity of three signs of infection: redness, pus, and swelling, and their skills were evaluated through repeated (quarterly) assessment of these images. Results were used to identify those CHWs needing more focused one-on-one training.
A sign-based algorithm  was followed by the CHWs to identify sick newborns and refer them for care. In the 33 clusters from the home-care arm of the previous research study, CHWs also had the capacity to treat newborns at home with injectable antibiotics, following procedures that have been previously described; these clusters were balanced across the three cord care regimens (11 clusters in each arm of this study, Table 1). At a final visit conducted between 28 and 35 days after birth, the CHW recorded the vital status of the baby.
Any neonatal death occurring during the period of follow-up was recorded; at regular intervals a database-generated list of recent deaths were distributed to supervisory staff responsible fro conducting an in-depth verbal autopsy. An automated algorithm utilized sign-based cause-of-death definitions to assign multiple and/or single cause of death to each infant, using both hierarchical and non-hierarchical assignment .
Outcomes, Sample Size, and Analysis
Eligibility and Enrolment Criteria
The two primary outcomes of this trial were neonatal mortality and omphalitis among live born enrolled infants. Neonatal mortality was defined as death within the first 28 days after birth and the sample size calculation (see Sample Size) was based on detection of mortality difference among enrolled infants. Omphalitis was defined under various sign-based algorithms representing mild, moderate, and severe omphalitis[11, 12]. Other a priori planned outcome analyses include 1) an assessment of the impact of the intervention on these outcomes among the entire sample of live born infants, including those that died before they had an opportunity to be enrolled in the trial, 2) analyses stratified by birth weight status (<2500 grams vs. > 2500 grams), gestational age (<37 weeks vs. >37 weeks), timing of initiation of intervention, and 3) analyses conditioned on survival to various time points (i.e. 24 hours, 48 hours, etc).
Total required number of newborns required for a range of detectable differences in mortality, by proportion of NMR occurring prior to enrolment
Detectable Difference in Mortality
Proportion of neonatal mortality occurring prior to enrolment
Data Management and Analysis
CHWs collected data using maternal-level forms during their routine area monitoring and the two pregnancy visits, and infant-level forms during their follow-up visits to the newborns. The VHW-collected pictorial information on intervention visits in the first 7 days of life (date, time, visit status) was confirmed by CHWs through discussion with family members and then transferred to the infant-level forms maintained by the CHWs. All forms were checked for completeness daily by CHWs and at fortnightly meetings with supervisory staff, and then transferred to Dhaka for entry into a customized Microsoft SQL Server database. Entry programs included standard range and validation checks. Selected process-related data variables were entered at the field level to enable rapid feedback on essential implementation variables (e.g. enrolment, coverage and timing of home visits, etc.) to the field level supervisory staff.
For the final (and interim) analyses descriptive statistics on enrolment rates, timing of intervention delivery, frequency and coverage of home visits, receipt of components of the basic package, reported care practices, and health and demographic characteristics of newborn, mother, and households will be examined for balance across the groups. The primary impact analyses will follow an intent-to-treat approach among all enrolled infants, and will focus on comparing the mortality experience of newborns across the three cord care regimen groups. The mortality rate per 1000 live births will be estimated for all groups, and risk of death in each of the intervention groups (multiple and single cleansing) relative to the dry cord care group will be calculated. Standard errors will be adjusted for the clustered allocation using the generalized estimating equation approach, and 95% confidence intervals will be estimated. If necessary, adjusted models will be constructed to account for any imbalance between the groups.
Individual signs of cord infection and omphalitis under three sign-based algorithms (mild, moderate, severe) will be compared between groups using similar regression models with adjustment for both clustered design and multiple measures per baby. Individual visit level cord assessment data will be collapsed over the neonatal period with binary indicators created for each of the sign-based algorithms to estimate cumulative incidence of cord infection and risk of omphalitis relative to the comparison dry cord group.
Data Safety and Monitoring Board and Interim Analyses
An independent Data Safety Monitoring Board (DSMB) consisting of national and regional experts was established. Interim analyses were conducted after 31.3% (March 2008) and 69.8% (January 2009) of the expected sample size was recruited. These data were presented to the DSMB in both electronic reports and face-to-face meetings in Dhaka, Bangladesh. Stopping guidelines for efficacy were agreed upon by the DSMB members and the investigating team prior to the trial, using Lan-DeMets boundaries and nominal p-values for group sequential analyses estimated using O'Brien/Fleming spending function[17, 18]. At each of the meetings, the DSMB recommended continuation of the project as originally planned.
The study protocol was approved by the Institutional Review Board of the Johns Hopkins Bloomberg School of Public Health and the Ethical Review Committee of the International Centre for Diarrheal Disease Research, Bangladesh (ICDDR,B). The protocol has been reviewed and re-approved on an annual basis since 2007.
Due to report in early 2010, the Projahnmo Chlorhexidine Study will provide an evaluation of the impact of multiple- or single cord cleansing with 4.0% chlorhexidine relative to the current WHO recommended dry cord care on neonatal mortality and omphalitis. The results from this trial will expand the evidence base for the use of topical chlorhexidine as a low-cost, simple-to-deliver intervention, and it's potential to reduce neonatal mortality and morbidity in low resource settings. Furthermore, it will provide new information on one-time cleansing of the umbilical cord and its potential effect relative to multiple cleansing. Since the chlorhexidine intervention was implemented in the context of a basic package of maternal and newborn care, this study will provide data on the impact of chlorhexidine cleansing which is additional to the effect of the basic package of care which is being scaled up in Bangladesh.
Rural Sylhet District of Bangladesh is an appropriate location for examining the impact of chlorhexidine interventions, with strong potential for generalizing results to the wider South Asian region. The rate of home births, overall neonatal mortality rates, and the proportion of deaths attributed to infection in this region are high and comparable to large parts of Bangladesh, India, Nepal and Pakistan. Access to care and care-seeking by households is low and identification of interventions that can improve survival and be delivered by community-based workers or caretakers in the household is urgently needed. The existing Projahnmo Project provided a highly capable and well-developed infrastructure for conducting a population-based community trial. Most importantly, Projahnmo represents a strong partnership that includes the Government of Bangladesh, local and international research institutes and NGOs. This broad-based partnership will increase the likelihood that if the intervention is found successful, it will be incorporated into national policy recommendations and scaled up within Bangladesh.
The data from this trial will be interpreted in conjunction with the results from the previous Nepal trial and will further inform ongoing discussions regarding current WHO recommendation for dry cord care. Further community-based trials are being planned for settings in sub-Saharan Africa, and lessons learned from the design and implementation of this trial might provide guidance to the conduct of those trials. If the intervention is ultimately proven successful, chlorhexidine cleansing of the cord can be rapidly incorporated into existing neonatal care promotion programs and contribute to global efforts to improve newborn survival.
List of Abbreviations Used
Community Health Worker
Neonatal Mortality Rate
Project to Advance the Health of Newborns and Mothers
Village Health Worker
World Health Organization
Funding for the Projahnmo Project is provided by the United States Agency for International Development, Office of Health, Infectious Diseases, and Nutrition, Global Health Bureau and the Dhaka Mission through the Global Research Activity Cooperative Agreement (GHS-A-00-03-00019-00), and the Saving Newborn Lives initiative of Save the Children Federation - USA through a grant from the Bill and Melinda Gates Foundation.
- Lawn JE, Cousens S, Zupan J, Lancet Neonatal Survival Steering Team: 4 million neonatal deaths: when? Where? Why?. Lancet. 2005, 365 (9462): 891-900. 10.1016/S0140-6736(05)71048-5.View ArticlePubMedGoogle Scholar
- World Health Organization: Care of the umbilical cord: A review of the evidence. WHO/RHT/MSM/98.4:,
- Mullany LC, Darmstadt GL, Tielsch JM: Role of antimicrobial applications to the umbilical cord in neonates to prevent bacterial colonization and infection: a review of the evidence. Pediatr Infect Dis J. 2003, 22 (11): 996-1002. 10.1097/01.inf.0000095429.97172.48.View ArticlePubMedPubMed CentralGoogle Scholar
- Zupan J, Garner P, Omari AA: Topical umbilical cord care at birth. Cochrane Database Syst Rev. 2004, 3 (3): CD001057-PubMedGoogle Scholar
- Mullany LC, Darmstadt GL, Katz J, Khatry SK, Leclerq SC, Adhikari RK, Tielsch JM: Risk of mortality subsequent to umbilical cord infection among newborns of southern Nepal: cord infection and mortality. Pediatr Infect Dis J. 2009, 28 (1): 17-20. 10.1097/INF.0b013e318181fb4c.View ArticlePubMedPubMed CentralGoogle Scholar
- Mullany LC, Darmstadt GL, Tielsch JM: Safety and impact of chlorhexidine antisepsis interventions for improving neonatal health in developing countries. Pediatr Infect Dis J. 2006, 25 (8): 665-675. 10.1097/01.inf.0000223489.02791.70.View ArticlePubMedPubMed CentralGoogle Scholar
- Denton GW: Chlorhexidine. Disinfection, Sterilization, and Preservation. Edited by: Block SS. 2001, Philadelphia, PA: Lippincott Williams & Wilkins, 321-326. 5Google Scholar
- Mullany LC, Darmstadt GL, Khatry SK, Katz J, LeClerq SC, Shrestha S, Adhikari R, Tielsch JM: Topical applications of chlorhexidine to the umbilical cord for prevention of omphalitis and neonatal mortality in southern Nepal: a community-based, cluster-randomised trial. Lancet. 2006, 367 (9514): 910-918. 10.1016/S0140-6736(06)68381-5.View ArticlePubMedPubMed CentralGoogle Scholar
- McClure EM, Goldenberg RL, Brandes N, Darmstadt GL, Wright LL, CHX Working Group, Armbruster D, Biggar R, Carpenter J, Free MJ, Mattison D, Mathai M, Moss N, Mullany LC, Schrag S, Tielsch J, Tolosa J, Wall SN, Schuchat A, Smine A: The use of chlorhexidine to reduce maternal and neonatal mortality and morbidity in low-resource settings. Int J Gynaecol Obstet. 2007, 97 (2): 89-94. 10.1016/j.ijgo.2007.01.014.View ArticlePubMedPubMed CentralGoogle Scholar
- Baqui AH, El Arifeen S, Darmstadt GL, Ahmed S, Williams EK, Seraji HR, Mannan I, Rahman SM, Shah R, Saha SK, Syed U, Winch PJ, LeFevre A, Santosham M, Black RE: Effect of community-based newborn care interventions package implemented through two service-delivery strategies in Sylhet district, Bangladesh: a cluster-randomised controlled trial. Lancet. 2008, 371: 10.1016/S0140-6736(08)60835-1.Google Scholar
- Mullany LC, Darmstadt GL, Katz J, Khatry SK, LeClerq SC, Adhikari RK, Tielsch JM: Development of clinical sign based algorithms for community based assessment of omphalitis. Arch Dis Child Fetal Neonatal Ed. 2006, 91 (2): F99-104. 10.1136/adc.2005.080093.View ArticlePubMedPubMed CentralGoogle Scholar
- Mullany LC, Faillace S, Tielsch JM, Stolzfus RJ, Nygaard KE, Kavle JA, Farag TH, Haji HJ, Khalfan SS, Ali NS, Omar RS, Darmstadt GL: Incidence and risk factors for newborn umbilical cord infections on Pemba Island, Zanzibar, Tanzania. Pediatr Infect Dis J. 2009, 28 (6): 503-509. 10.1097/INF.0b013e3181950910.View ArticlePubMedGoogle Scholar
- Darmstadt GL, Baqui AH, Choi Y, Bari S, Rahman SM, Mannan I, Ahmed AS, Saha SK, Rahman R, Chang S, Winch PJ, Black RE, Santosham M, El Arifeen S, Bangladesh Projahnmo-2 (Mirzapur) Study Group: Validation of community health workers' assessment of neonatal illness in rural Bangladesh. Bull World Health Organ. 2009, 87 (1): 12-19. 10.2471/BLT.07.050666.View ArticlePubMedGoogle Scholar
- Baqui AH, Arifeen SE, Williams EK, Ahmed S, Mannan I, Rahman SM, Begum N, Seraji HR, Winch PJ, Santosham M, Black RE, Darmstadt GL: Effectiveness of home-based management of newborn infections by community health workers in rural Bangladesh. Pediatr Infect Dis J. 2009, 28 (4): 304-310. 10.1097/INF.0b013e31819069e8.View ArticlePubMedPubMed CentralGoogle Scholar
- Baqui AH, Darmstadt GL, Williams EK, Kumar V, Kiran TU, Panwar D, Srivastava VK, Ahuja R, Black RE, Santosham M: Rates, timing and causes of neonatal deaths in rural India: implications for neonatal health programmes. Bull World Health Organ. 2006, 84 (9): 706-713. 10.2471/BLT.05.026443.View ArticlePubMedPubMed CentralGoogle Scholar
- Zeger SL, Liang KY, Albert PS: Models for longitudinal data: a generalized estimating equation approach. Biometrics. 1988, 44 (4): 1049-1060. 10.2307/2531734.View ArticlePubMedGoogle Scholar
- Lan KK, DeMets DL: Discrete sequential boundaries for clinical trials. Biometrika. 1983, 70: 659-663. 10.2307/2336502.View ArticleGoogle Scholar
- O'Brien PC, Fleming TR: A multiple testing procedure for clinical trials. Biometrics. 1979, 35 (3): 549-556. 10.2307/2530245.View ArticlePubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2431/9/67/prepub