Neonatal Survival in Complex Humanitarian Emergencies: Setting an Evidenced-Based Research Agenda Dr. Basia Tomczyk Dr. Diane Morof CDC IAWG Annual Meeting May 31-June 1 Kuala Lumpur, Malaysia

Outline Background Goal Methods Findings Limitations Conclusions

Background Saving Newborn Lives (SNL) developed a list of questions deemed critical to improving the evidence base of research. STC selected those that could potentially be done in a humanitarian context and that are most likely relevant to humanitarian settings. 36 questions selected.

Background (cont’d) Systematic method for setting priorities in health research investments Multiple step process Specify context Discuss and choose criteria Select questions and apply criteria to the questions Flexible process enables prioritization at any level: institutional, regional, national, international or global Used in child health and nutrition, RH research workshop, mental health and psychosocial support in humanitarian emergencies Rudan et al. Setting Priorities in Global Child Health Research Investments: Guidelines for Implementation of CHNRI Method. Croat Med J. 2008; 49:720-33. www.cmj.hr

Research Goal The goal was to adapt the Child Health Research and Nutrition Initiative (CHRNI) methodology to prioritize neonatal health research in the unique setting of complex humanitarian emergencies.

Methods The context was complex humanitarian emergencies   The time was the interval from now to year 2020   The population disease burden of interest was all mortality and disability due to neonatal causes   Survey monkey

Criterion 1: Answerability Definition: Likelihood that research would lead to new knowledge in an ethical way   Would you say the research question can be well framed and endpoints can be well defined?     Would you say that local and national research capacity already exists to answer the research question in many different contexts? Do you think that a study needed to answer the proposed research question would obtain ethical approval without major concerns?              

Criterion 2: Feasibility Definition: Likelihood that research can be conducted cost-efficiently and lead to generalizable new knowledge     Would you say that well designed studies of moderate scale and sample size would have enough power to provide replicable answer to proposed research question?       Would you say that ongoing programs, resources and research capacities could be amended or expanded to undertake proposed research or uptake the measurement technique cost-efficiently? Would you say that the answer to proposed research question would be generalizable across many (or most) different contexts?                      

Criterion 3: Relevance Definition: Likelihood that research would address important condition and critical gap in knowledge and could be readily translated to inform policies and programs     Is the condition being assessed a high burden condition?         Is the proposed research addressing a critical gap in knowledge? Would you say that the results of proposed research could be readily translated/used in health information systems, to improve service delivery, to inform policies and programs?          

Criterion 4: Equity Definition: Likelihood that the proposed epidemiological research would have positive impact on equity and local ownership In given context, would you say that the present distribution of disease burden primarily addressed by the proposed research affects IDPs, refugees, and communities affected by humanitarian crisis? Would the proposed research and its results be owned by local actors (e.g. district managers, health workers, communities)? Would you say that the proposed research has the overall potential to improve equity in disease burden distribution in the longer term (e.g. by 2020)?      

Methods (cont’d) Analysis: Apply criteria to the research questions Yes’s were given 1 point, no’s zero and don’t know 0.5 Determined research priority score (RPS) with a cut off of 0.65 Same list of initial agencies offered list of questions 36 agencies and individuals. Criteria applied to triage questions by answering Yes, no, don’t know. Used online survey zooomerang to collect data. We followed a similar methodology to the two articles sent out including the Setting Implementation Research priorities to reduce preterm births and stillbirths at the community level (George et al) January 2011 to determine an overall research priority score. Yes’s were given 1 point, no’s zero and don’t know 0.5. 12 responses some on behalf of agencies Respondents were then asked to Weighting the importance of the criteria. Sent to 20 agencies and received 6 responses. Based on the importance given each criteria was given a weight, and the RPS was weighted. We used 0.65 cutoff for the RPS In addition, to better understand if there was consensus on the answers. A average expert agreement score was generated. The AEA was calculated based on the average most frequent response for all the criteria.

Results 28 respondents completed 38 began but didn’t complete Research prioritization scores ranged from 0.846 to 0.679

Results (cont’d) Four questions dealt with the following: What is the additional burden of neonatal mortality? Can pregnancy surveillance be used to measure neonatal mortality? Can verbal autopsy be used to capture causes of neonatal mortality? What are the risk factors for neonatal sepsis?

Results (cont’d) The programmatic priorities included a focus on: How can we increase demand for skilled attendance at birth? What is the coverage of clean delivery practices? What are the factors that can increase facility delivery uptake? What are the best strategies to identify preterm babies at the community level?

Limitations Question clarity, intent, and feasibility Criteria interpretation Uncertain if we reached appropriate participants Time Very technical questions, unclear on some what the intent was and some seemed unlikely to be feasible in humanitarian settings Also for the conference we are focusing on acute emergencies but the survey was not based on that. Because we sent it to the IAWG listserve. Criteria could be interpreted buy individuals

Conclusions The research priorities identified recognize a need to focus on both the descriptive epidemiology and operations research to improve neonatal health in these settings. The CHRNI exercise has already stimulated discussion and action to address some of the research gaps by donors, international nongovernmental organizations (NGOs) and researchers.

Thank you!