Digital technology, including the internet, smartphones, and wearables, provides the possibility to bridge the mental health treatment gap by offering flexible and tailored approaches to mental health care that are more accessible and potentially less stigmatising than those currently available. However, the evidence base for digital mental health interventions, including demonstration of clinical effectiveness and cost-effectiveness in real-world settings, remains inadequate. The James Lind Alliance Priority Setting Partnership for digital technology in mental health care was established to identify research priorities that reflect the perspectives and unmet needs of people with lived experience of mental health problems and use of mental health services, their carers, and health-care practitioners. 644 participants contributed 1369 separate questions, which were reduced by qualitative thematic analysis into six overarching themes. Following removal of out-of-scope questions and a comprehensive search of existing evidence, 134 questions were verified as uncertainties suitable for research. These questions were then ranked online and in workshops by 628 participants to produce a shortlist of 26. The top ten research priorities, which were identified by consensus at a stakeholder workshop, should inform research policy and funding in this field. Identified priorities primarily relate to the safety and efficacy of digital technology interventions in comparison with face-to-face interventions, evidence of population reach, mechanisms of therapeutic change, and the ways in which the effectiveness of digital interventions in combination with human support might be optimised.

Patient and public involvement in health research is increasingly well established internationally, but the impacts of involvement on the research process are hard to evaluate. We describe a process of qualitative data analysis in a mental health research project with a high level of mental health service user and carer involvement, and reflect critically on how we produced our findings. Team members not from research backgrounds sometimes challenged academic conventions, leading to complex findings that would otherwise have been missing. An essential component of how we coproduced knowledge involved retaining methodological flexibility so that nonconventional research voices in the team could situate and critique what was conventionally known. Deliberate and transparent reflection on how "who we are" informed the knowledge we produced was integral to our inquiry. We conclude that reflecting on knowledge (co)production is a useful tool for evaluating the impact of patient and public involvement on health research.

The search for an optimum model of involvement may prove elusive, but the need to research and debate different strategies, to avoid tokenism and exploitation, remains.