Group Reports & & & & & & & & & & Journal of Perinatology ( 2002 ) 22, S27 – S32 doi:10.1038/sj.jp.7210807 The participants divided into four groups on days 3 and 4 of the workshop to discuss and prepare recommendations on the following four areas: Components of community-based interventions to improve perinatal and newborn outcomes? Appropriate study designs and key variables for studies on community-based interventions to improve perinatal and newborn outcomes? Steps and factors in moving from research to implementation? Eliciting community mobilization and focused community participation to promote neonatal health? & & Type of intervention 1.1 Maternal nutrition Iron and folate Iodine Balanced energy – protein supplementation Zinc Zulfiqar Bhutta, Robert Black, Ruth Frischer, Joseph de Graft-Johnson, D. S. Manandhar Vitamin A 1. Components should be linked to existing community-based Safe Motherhood initiatives and child-survival programs. 2. Interventions should be based on bedrock of scientific evidence, cost-effectiveness and programmatic sustainability. 3. Interventions within existing programs that lead to improved infant and child health or have long-term benefits should continue. The components of the intervention package were classified as essential (to be always included) or desirable (to be included whenever feasible). The group reviewed and recommended potential interventions in a conceptual framework of antenatal, intrapartum (including the immediate postpartum period), and postnatal care. & 1. Antenatal Care GROUP REPORT 1 Components of community-based interventions to improve perinatal and newborn outcomes. The purpose of several evolving research studies is to determine how to deliver a package of cost-effective and sustainable perinatal–neonatal health interventions in the community. In order to assemble a potentially effective and ‘‘complete’’ package, it is essential to examine the available interventions and prioritize them. This group was assigned the task of identifying the components of such a package. The group recommended the components of the package of interventions on the basis of the following conceptual framework: & Multivitamins Justification and research gaps Recommendation Strong evidence for maternal benefits Essential ( pre - , during and postpregnancy ) Essential ( pre - , during and postpregnancy ) Strong evidence of benefit in areas of endemic iodine deficiency Some evidence of benefit in research settings. Issues of sustainability and operational implementation remain. Need for further research on benefits in diverse settings. Need for further research Need for further research Need for further research 1.2 Prevention and treatment of infections Treatment of Need for further syphilis evaluation of burden and impact of interventions in developing country settings Strong evidence of Treatment of malaria maternal benefits. ( intermittent ) Some evidence of benefits to the newborn. Merits further research in malaria endemic areas. Need for further work Prevention of on cost - effective mother - to - child interventions transmission of HIV Need for further Treatment of research in developing symptomatic countries urinary tract infection Treatment of Need for further asymptomatic research in developing bacteriuria countries Need for further evaTreatment of luation of burden and other sexually impact of interventions transmitted in developing country disease ( e.g., settings chlamydia, gonococcal infections ) Desirable in areas with existing food supplementation programs. Essential in endemic areas Essential in endemic areas Essential in HIV endemic areas Desirable (continued on next page) Journal of Perinatology 2002; 22:S27 – S32 # 2002 Nature Publishing Group All rights reserved. 0743-8346/02 $25 www.nature.com / jp S27 Group Reports (continued ) Type of intervention Detection and treatment of bacterial vaginosis 1.3 Behavioral issues Birth preparedness Education into seeking antenatal care Counseling on benefits of exclusive breastfeeding Improved nutrition in pregnancy Recognition of danger signs for seeking emergency obstetric care Education on high risk pregnancies, e.g., previous stillbirths or perinatal deaths (continued ) Justification and research gaps Recommendation Need for further research Cord cutting and care Essential Essential Essential Essential Training of the birth attendant and families in recognition of birth complications that impact on newborn health and require speedy referral, e.g., malpresentation, prolonged labor, eclampsia, antepartum hemorrhage, cord prolapse Hand - washing by birth attendant, clean surface Care of the baby during the delivery of the placenta  Drying and wrapping the baby  Placing the baby on the mother’s abdomen or to the breast S28 Use of clean disposable blade  Use of clean tie  Clean cord and leave uncovered Basic resuscitation including: Justification and research gaps Recommendation Further operational research is needed to assess the utilization of safe birth kits versus other approaches, and the impact of cord antisepsis on neonatal outcomes Essential Need for further operational research, especially at the domiciliary and community level Essential Need for behavioral and operational research Proven benefits Essential  Essential Tactile stimulation Positioning  Mouth - to - mouth or bag - mask resuscitation Avoidance of inappro priate oxytocic use   Need for further research on benefits for perinatal / newborn outcomes Desirable Immediate breast feeding 2. Intrapartum Care and Immediate Postpartum Care The group underscored the importance and the need for skilled health care at birth, but recognized that the nature of individuals providing this care would differ according to health systems and countries. The emphasis should be on the requisite skills necessary for safe birth in domiciliary settings. Type of intervention Type of intervention Justification and research gaps Recommendation Well - known complications that impact on fetal – neonatal survival Essential Essential Need for research on the optimal timing of cord cutting in relation to placental delivery Essential Essential 3. Postnatal Care Type of intervention Keeping the baby warm  Delay bathing by 6 hours or more  Kangaroo Mother Care  Swaddling  Co - bedding  Warm room Weighing the baby Exclusive breast - feeding Cup and spoon feeding of expressed breast milk for low birth weight infants Hygiene and skin care Eye care Justification and research gaps Recommendation Need for operational research Essential Need for research on alternative methods ( e.g., anthropometry ) to identify low birth weight babies Proven benefits Well - established benefits Desirable Need for formative and operational research Proven benefit Essential Essential Essential Essential in endemic areas Journal of Perinatology 2002; 22:S27 – S32 Group Reports (continued ) Type of intervention Education of families and caregivers for recognition of danger signs on care seeking and referral of a sick newborn Education of families and caregivers in recognition of danger signs for neonatal sepsis and care seeking / referral / treatment Treatment of neonatal sepsis Justification and research gaps Need for further research into use of a clinical algorithm to identify sick newborns, domiciliary management of the sick newborn, if referral not possible. Operational research also needed on how to overcome barriers to identify sick newborns and how to promote effective referral pathways Need for research to validate criteria of sepsis in diverse settings. Need for research on microbiology of organisms causing sepsis Several possibilities exist for treatment of sepsis Selection of antibiotics should be based on microbiology of organisms, national drug policy, cost effectiveness and feasibility of delivery. Need for research into different approaches for treating neonates with sepsis in the community, including domiciliary management. Recommendation Essential Essential Key Outcome Indicators The choice of outcome indicators depends on the duration and size of the study. These two factors affect the likelihood of achieving changes in outcome indicators such as neonatal mortality rate. The indicators that are expected to change due to intervention should be specified at the outset. If a process indicator is used as an outcome, it should also be used in the estimation of sample size. For studies on community-based interventions to improve perinatal and neonatal health, the following indicators were recommended: Essential Note: The group highlighted the need to consider approaches to domiciliary management of the newborn in the context of the health system. Thus, it is important to emphasize the need to strengthen referral pathways, improve quality of care at the primary and secondary referral centers, and ensure linkage with maternal postnatal care. GROUP REPORT 2 Saadet Arsan, Paul Arthur, Abdullah Baqui, Simon Cousens, Ashok Deorari, Shams El Arifeen, David Osrin, Izaz Rasul, Indira Narayanan Appropriate study designs and key variables for studies on community-based interventions to improve perinatal and neonatal outcomes. In order to maximize the gains from different studies, it is essential to have standardized design and methodology. Only then can the results be compared and, possibly, pooled. There should be an agreement on key outcome variables and date collection methods: This group addressed these key issues. Journal of Perinatology 2002; 22:S27 – S32 Study Design The optimal study design to demonstrate the impact of an intervention on a behavior or a health outcome is the randomized controlled trial. Current knowledge about community care of neonates is not at a stage where intervention packages can be taken to trial without the need for randomization and controls. The use of controls raises ethical and design issues. One way to address these concerns is to introduce interventions in a stepwise fashion. The packages of interventions should be examined for effectiveness, while individual interventions need to be tested for efficacy.  Essential: Neonatal mortality rate  Desirable: Perinatal mortality rate, stillbirth rate, early neonatal mortality rate, late neonatal mortality rate, fresh stillbirths plus neonatal mortality rate. It was recommended that cost be estimated, as it is essential descriptive information in these studies. Depending on the scope of the study and the precision of the evaluation tools employed, the following additional information may be gathered: Maternal morbidity Fresh stillbirth/all stillbirths Cause-specific neonatal mortality Birth weight (or surrogates) Proportion of low birth weight infants Birth-weight-specific mortality Proportion of preterm infants Incidence of neonatal morbidities, especially birth asphyxia and sepsis  Delivery sites, skilled or unskilled care at birth  Choice and use of providers, utilization of health system resources         Process Indicators One or more process indicators could be used as outcome variables. Generating information on practices is more important than information on knowledge; and observation or demonstration of activities is preferable to reporting or recall of activities. Qualitative assessment must be an integral part of the studies as ethnographic investigation may yield extremely useful information for the program. S29 Group Reports The specific process indicators recommended were skilled health care at birth, immediate care at birth, thermal control, clean practices at birth, resuscitation, cord care, early initiation and exclusivity of breast-feeding, eye care, immunization, recognition of danger signs, care seeking, and postnatal care by a skilled provider. Many of these indicators need further definition and validation. Considerations in Data Collection Data collection may be continuous or periodic. Both have their advantages and disadvantages. A decision needs to be taken beforehand whether specific information will be collected from the control group or not. The process of data collection may affect the outcomes in the control population. The greater the quantum of process information collected, the higher the cost of the study and the more intense the effect on the control group. One way to neutralize this effect is to exclude informants from subsequent periodic assessments. As far as possible, studies should develop their own evaluation system and not rely on routinely collected data. Studies need to take into consideration the issues of loss to followup, variable exposure to intervention, dispersion or congruity of intervention and control areas, intra- and intercluster movement, and problems of emigration or temporary movement of subjects. All providers within the community should be enumerated. Standard and accepted definitions and tools should be used. The studies may address the development of additional necessary tools as a part of their activities. GROUP REPORT 3 Elaine Albernaz, Abhay Bang, Gary Darmstadt, Judith Moore, N. C. Saxena, Uzma Syed Moving from research to implementation. The ultimate aim of action research is to develop interventions for incorporation into health programs. However, all community-based research does not necessarily lead to program-ready interventions and approaches. This group was assigned the task of defining the attributes of interventions that make them amenable to implementation in the program setting and the steps required to achieve that. Criteria The decision to move from research to implementation depends on several considerations: 1. Is the problem of sufficient magnitude and/or severity in the population to take up the intervention on a programmatic scale? The importance of a given health problem varies from country to country. There is no universal threshold of neonatal mortality rate at which a program targeted to newborn infants needs to be initiated. Each country should set national goals for neonatal health indicators. S30 2. Is the intervention effective? The intervention should have been demonstrated as effective under field conditions in more than one site and under more than one set of conditions. There is often a need for an intermediate step of effectiveness studies between the initial efficacy trial and implementation. The intervention should potentially be able to address the problem on a national level. The acceptable level of impact depends on many factors such as the cost, the ease of implementation and potential for adverse effects. 3. Is the intervention cost-effective? The intervention should be cost-effective and affordable by the country. 4. Is the intervention acceptable? It is essential to ensure the acceptability of the intervention among decision makers, professional leaders, health workers and the community. 5. Is the intervention sustainable? Sustainability is a fundamental issue in community programs. Embarking on a program that cannot be sustained in the long run does more harm than good to the cause. Vertical programs may be more easily scaled up than an integrated, multicomponent program, but sustainability may be compromised. 6. Is there an ethical urgency? There is a threshold beyond which it is unethical to withhold an intervention from the community that has impact and only a limited chance of causing adverse effects. Steps After the above criteria regarding the intervention are met, the following steps are required in order to effectively implement the intervention in a program:  Build consensus among political leaders and other stakeholders that implementation of the intervention is essential. Generating political will and momentum may be a major bottleneck to the launch of a program. Therefore, there is a need to take along all the stakeholders including national and regional policy makers, professional groups, human rights groups, women’s groups and the media. Opinion makers who are likely to oppose the program should be included in deliberations from the beginning.  Select the implementing agency, which could be a department within the government, private sector, or an NGO, or a combination of these.  Constitute a group that adapts the intervention to national or regional circumstances. It is possible that some elements of the package are dropped or modified.  Prepare a blueprint for implementation of the intervention, taking into account the burden of the problem, perceived community demand for the intervention, resource needs, health system strengths/weaknesses, private sector/community assets and available tools. Journal of Perinatology 2002; 22:S27 – S32 Group Reports  Start the phase-wise implementation of the program. Initial communities should be selected carefully to provide for a fair chance of success of the program.  Monitor and evaluate the program. There is a need to regulate the implementation of the program to ensure that the program improves over time. Aspects of the package may lose relevance over time as the magnitude of the target problem(s) changes, requiring fine-tuning of the program. The above process will need to be repeated at the local level to make microlevel adjustments. health because infants are often born at home and the care of newborns in developing countries is often conditioned by traditional beliefs, practices, and behaviors that may require change for improved outcomes. The deliberations of this group were focused on the need, principles, and methodology of community mobilization. Why Community Mobilization and Participation  Focus of the program is on the infants of the community; hence, its involvement is a moral and ethical obligation.  Newborn essential care packages consist primarily of GROUP REPORT 4 Nabeela Ali, Anthony Costello, Lisa Howard-Grabman, Vinod K. Paul, Jose Martines Eliciting community mobilization and focused community participation to promote neonatal health. Community acceptance and participation in health care programs is a fundamental prerequisite for its sustainability and success. Investigators engaged in community-based research aimed to be translated into programs must build in a component of community mobilization from the outset. This is particularly so for neonatal     behavior change, requiring changes in individual, family, and community behaviors and norms. Success of community-based interventions depends on community acceptance of the interventions. Involvement of communities in research planning, implementation of interventions and evaluation can help to reduce lag time for implementation, upscaling, and expansion. Communities can contribute resources that may not be available to the health system alone. Community involvement in monitoring and evaluation provides community perspective on quality of care, appropriateness, and feasibility of interventions. Figure 15. The community action cycle for community mobilization (Courtesy, Lisa Howard-Graham ). Journal of Perinatology 2002; 22:S27 – S32 S31 Group Reports  Community participation breaks barriers to implementation,  Consider issues of coverage of all strata including those most helps program teams to troubleshoot, and develop practical solutions.  Outcomes of community participation go beyond health to include self-efficacy and collective efficacy, and strengthened community capacity to identify priorities and address them.  Community participation is the key to long-term sustainability of program interventions and outcomes. affected by the problem with emphasis on marginalized or disadvantaged groups. Ensure appropriate involvement of ‘‘gatekeepers,’’ opinion leaders, and decision makers (including private practitioners). Develop strategies to ensure continuing community interest in the initiative (e.g., identifying successes and celebrating them). Ensure that the team has a clear understanding of economic, social, political, and cultural context. Work in accordance with the community calendars and schedules. Make the program management culture as participatory and open to supporting new approaches. Do not punish team members for mistakes; learn from them. Keep focused on the program goal; put a monitoring system in place with community review of findings. Ensure adequate contact with the community to maintain desired level of action. Determine your program policy and philosophy with regard to incentives for community participation in the context of long-term sustainability of the program and not for shortterm gains.    Some Basic Principles of Community Participation/ Mobilization  Set specific goals around which to mobilize the community.  Define your community (it could be just a ‘‘core group,’’ or   supporting groups and individuals, or a broader community).  Be apolitical. Do not indulge in local politics.  Be transparent and honest; explain the goals of the program  to the community and share what you can do and what you cannot do. Respect community values and concerns as well as hierarchies and culture. Define and redefine your role in relation to communities as they strengthen their capacity in order to encourage growing autonomy. If you do not know why something is happening, ask community members. Be flexible. Not all communities are the same and not all groups within communities are the same. Define process objectives for community mobilization related to underlying themes that contribute to the health problem (e.g., exploring the value of newborns’ and mothers’ lives, gender equity, shared responsibility for quality care). Draw upon successful models in similar situations.        Community Action Cycle Community action cycle describes the steps in community mobilization. It consists of the following sequence of activities: prepare to mobilize, organize community for action, explore the health issues and set priorities, plan together, act together, evaluate together and scale up. Figure 15 depicts the details of these steps.  Issues in Scaling Up  The methodology needs to be affordable and feasible in order to be scaled up.  Consensus should be built around adopting a particular model/approach.  Partnerships are essential to achieve larger scale impact and it is crucial to define goals of the partnership and program, roles and responsibilities, coordination mechanisms, and indicators of success/performance.  Partners in scaling up could be social scientists, NGOs, government, communities, private sector, economists, and other technical specialists.  Advocacy with the leaders is required to ensure acceptance of the scaling-up initiative. Gaps Issues in Implementation of a Community Mobilization Effort  There are only a few examples of scaling-up experiences and  Ensure that design is feasible and affordable on a large scale.  Develop capacity of facilitators for community mobilization community participation programs at different scales and in different settings.  Only a limited experience is documented on the impact of community participation approaches combined with communication strategies such as social marketing, mass media, and interpersonal communication and counseling.  While estimating cost-effectiveness, the investigators often overlook gains in community capacity as a part of the broader benefit to the program. through a combination of training and field experience. (This is a technical discipline with a defined set of skills.)  Foster community acceptance of the project and its team at the outset.  Build on other existing programs and activities in the community.  Build on prevalent positive practices in the community. S32 approaches to scaling up in community mobilization.  There is a paucity of research on impact evaluation of Journal of Perinatology 2002; 22:S27 – S32