Process evaluations systematically assess the implementation of a programme or intervention and explore the mechanisms through which intended or unintended outcomes and impacts are produced. They also interrogate the broader context within which delivery occurs, recognising that context can both influence and be influenced by implementation processes and intervention effects. Process evaluations are critical for understanding not only whether an intervention works, but also how, why, and under what conditions.

Drawing on the MRC framework for process evaluation of complex interventions, we focus on three core domains to structure process evaluations: Implementation, Mechanisms of Impact, and Context. 

Implementation

The implementation domain focuses on both the process of delivery ("how") and the content of delivery ("what"). 

Specifically, the process of delivery examines "how the intervention was delivered", including: 

• Delivery structures, resources, and any modifications made during implementation;

The content of delivery focuses on "what was delivered", encompassing:

• Fidelity: the extent to which delivery adhered to the original design;
• Dose: the amount of intervention exposure received by participants;
• Reach: the proportion of the intended target population that engaged with the intervention;
• Adaptations: deviations or modifications made during implementation, whether planned or unplanned. 

For example, in a public health programme promoting the uptake of family planning among women in rural areas, a process evaluation may assess whether educational sessions were delivered in accordance with programme specifications (fidelity), the number of sessions each participant attended (dose), and the extent to which women from the target population were engaged (reach). 

Mechanisms of Impact

Mechanisms of impact refer to the processes through which intervention activities generate change. Process evaluations explore:

• Participant responses to and interactions with intervention components, including perceptions of acceptability and satisfaction;
• Mediators, or the intermediate processes that link intervention activities to outcomes, such as increased knowledge, changes in attitudes, or strengthened self-efficacy;
• Unanticipated pathways and consequences, encompassing both positive and negative outcomes not originally envisaged in the programme design. 

Continuing the family planning example, participant trust in health workers may emerge as a key mediator influencing the uptake of contraceptive methods. Conversely, unanticipated consequences might include the entrenchment of misconceptions if programme messages are misinterpreted. 

Context

The context domain captures the environmental, institutional, and socio-cultural factors that shape and are shaped by the intervention. It includes:

• Theoretical assumptions underpinning the intervention and the extent to which they align with or challenge prevailing local conditions;
• The influence of context on intervention implementation, for example, whether institutional commitment, resource availability, or frontline worker capacity facilitate or constrain delivery;
• Contextual causal mechanisms, whereby external factors—such as cultural norms, religious beliefs, gender relations, or political dynamics—interact with intervention processes to produce variation in outcomes across different settings. 

For example, in the context of family planning, local gender norms and religious teachings concerning contraception may significantly affect intervention acceptability and success, independent of the technical fidelity of delivery.

Further details about the current designs, methods, and resources in process evaluation that we utilise and focus on are available below.