Impact Evaluation of Disability Inclusiveness Graduation (DIG) in Uganda

This study will investigate the impact of BRAC’s Ultra Poor Graduation (UPG) program on poverty, livelihood, food security, and social participation outcomes of people with disabilities and their families in Kiryandongo, Gulu, Nwoya and Oyam districts of Uganda. It also aims to estimate the different impacts of the intervention between clients with and without disabilities. In addition, it will help to generate knowledge about the effectiveness of the UPG program on people with disabilities so as to inform future interventions.

Researchers: Professor Hannah Kuper; Dr Munshi Sulaiman; and Christine Nabulumba

Partners: The London School of Hygiene & Tropical Medicine (LSHTM)

Timeline: 2020–2022

Status: Ongoing

Contact: Christine Nabulumba


This disability-inclusive graduation model will have four main components to address the needs of people with disabilities. The first one is livelihood intervention, which includes the transfer of assets, technical training, and individual-level support for income generation. The next one, social protection, will include a six-month consumption stipend, healthcare subsidy and rehabilitation, physiotherapy and psychosocial support. Financial inclusion will cover financial literacy training, village savings and loans association (VSLA) formation as well as continuous coaching. Finally, social empowerment will be done through home coaching to provide individual counselling and life skills, and individual empowerment plans.

Research Questions

  1. What is the impact of the program on poverty, livelihood and social participation of people with disabilities and their families?
  2. How does the impact differ between people with disabilities compared to people without disabilities who received the intervention?


The study follows a cluster randomized control trial (RCT) method, which will include village clusters. Eligible project participants (5,300) in eight BRAC branch offices in the Kiryandongo, Gulu, Nwoya and Oyam districts of Uganda will be randomized. First, from the 156 villages, 185 artificial clusters were created, each with at least 10 eligible households. To create a cluster, either nearby small villages were grouped or larger villages were split to create a manageable cluster, that is, having between 10 and 75 eligible households. Second, within each BRAC branch, each cluster was randomly allocated to either treatment or control until a DIG-targeted number of beneficiaries was achieved. In the end, 96 were assigned to the treatment and the remainder 89 clusters to the control group. Overall, 2,898 households were assigned treatment while 2,402 households were retained as control. Third, while all the 185 clusters were included in the study, a household inclusion depended on whether they had a person with a disability or not. Households with persons with disability were automatically included in the study while households without persons with disability were included on a 60% probability per cluster. In total, 1,012 households with a person with disability and 2,644 households without a person with disability were included in the study.


Study ongoing.