Evidence in the face of adversity: studying psychological interventions in humanitarian settings
February 15, 2017
Conducting high-quality research in humanitarian settings is challenging but possible. We must find ways to meet these challenges and do this research.
Communities affected by adversity, including those facing challenges caused by major humanitarian emergencies, need effective psychological interventions, but there is a lack of practical and available tools to help large numbers of people. The World Health Organisation (WHO) is working to fill this gap by developing and testing potentially scalable psychological interventions that are specifically designed for such communities. The work builds on successes of the last 15 years by researchers who have evaluated psychological interventions in low and middle-income countries. We are trying to take this work one step further by simplifying the interventions where possible and making them publicly available if proven effective. The assessment of their effectiveness will need to cope with the many challenges of conducting high-quality research in humanitarian settings, but we know that this is possible. In this blog, we describe what WHO is doing and how it may meet the needs of the affected communities, those who are trying to help them and the research and policy actors who require reliable and robust evidence.
The WHO interventions are typically modified from existing evidence-based psychological interventions, with the aim that they will reduce the burden on scarcely available specialists. These modified interventions can be delivered over fewer sessions, by a non-specialist or provided as self-help. They include group and individual face-to-face non-specialist delivered interventions, e-mental health interventions and a multi-media intervention involving both audio recording and self-help reading materials. Before releasing the materials globally as WHO-recommended interventions, each intervention will be typically tested through at least two high-quality pragmatic randomized trials, to ensure that it is effective and safe. The randomized evaluation of the first of the interventions, Problem Management Plus, was funded by R2HC, and published in JAMA.
We – in strong partnership with a range of humanitarian agencies and universities – are testing a variety of psychological interventions in Pakistan, Lebanon, Uganda, Nepal and Kenya. When testing in these locations, we consider research design elements carefully to ensure a good quality evidence base while maintaining ethical standards. We invest in sensitive and comprehensive translation and contextual adaptation of intervention materials and use a formative research phase to ensure understanding, acceptability, and meaningfulness of the intervention. This can include literature reviews, rapid community qualitative research, blind back translations, and interviews with local people. It can be particularly difficult when adapting text-based materials into a language that has no written form, such as Pashtun or Juba Arabic.
Lack of resources must be carefully considered and the research design must be flexible to the study environment. For example, planning for provision of Wi-Fi and tablet stations in primary health centres to increase accessibility of e-health interventions may prove essential in an upcoming evaluation of an e-mental health intervention in Lebanon. While the fidelity checks needed for a high quality randomised trial may require live peer-provided checks rather than recordings in areas where participants are suspicious of recording procedures. Alternative contact methods for follow-up interviews to assess the impact of the intervention will also be needed in communities with limited access to phones, and research assistants have gone to the homes of participants or got permission to call a neighbour’s phone in studies in Kenya and in a South Sudanese refugee camp in Uganda.
Researchers also need to become accustomed to flexible timelines that may be needed to account for delays due to unexpected events that are more likely in humanitarian settings than in the relative calm and order of other research settings. We saw this in Kenya where flooding and transport issues created a barrier to recruitment for some weeks, and in Pakistan where vaccination campaigns were a public health priority and hindered data collection for our trial. Working with people who are vulnerable through poverty also makes the management of participant expectations especially important. Informed consent procedures need to make it clear to participants that the interventions will not provide material or financial benefits.
Conducting high-quality research in humanitarian settings is challenging but possible. We must find ways to meet these challenges and do this research if we want to make sure that greater numbers of affected people can be helped in effective ways.
Melissa Harper Shehadeh is a consultant working on global mental health at the World Health Organization and completing a PhD in Global Health at the University of Geneva.
Mark van Ommeren is a Public Mental Health Adviser at the Department of Mental Health and Substance Abuse at the World Health Organization. He has been working in the field of mental health and humanitarian settings for more than 20 years.
Photo: Copyright WHO – M. Kokic.
This blog was originally posted on Evidence Aid.